Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018 FIFA World Cup

Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018... We conducted an audit experiment to examine whether street vendors in Bogotá (Colombia) exert price discrimina- tion based on buyers’ attributes, such as gender and nationality, and based on product characteristics, such as the increasing marginal valuation of items needed to complete a collection. We exploited the seasonal demand for album stickers related to the FIFA World Cup Russia 2018. In our within-subjects design, experimenters carried out in-person audits and quoted a pre-determined list of missing stickers. They interacted with 59 sticker vendors located in five geographic clusters and collected 287 vendor–buyer interactions. We find that prices quoted to foreign buyers are higher than prices quoted to Colombian buyers. By contrast, we do neither find evidence supporting direct gender- based discrimination, nor that vendors charge a higher price per sticker when the list of missing stickers is shorter. We complement the study with a qualitative analysis based on interviews that reveal vendors’ pricing strategies, their awareness of price discrimination, and the trade of counterfeits. The qualitative results suggest that price discrimina- tion appears to be unconscious. Keywords: Colombia, Dual labor markets, Football, Latin america, Sports, Street vendors JEL Classification: C93, J46 1 Introduction For a season related to a sports competition, such as the FIFA World Cup, which we explore in this study, vendors The coexistence of formal and informal labor markets is prefer to have an accelerated turnover of merchandise prevalent in low and middle-income countries. In par- instead of storing it, since it is a very costly option. There ticular, street vending accounts for a significant part of - their informal labor sector since it is a major source of fore, in the absence of posted prices in informal markets, employment and income in the main cities. Some street we explore whether price discrimination appears as a vendors sell the same products year-round, while oth- possible mechanism, either conscious or unconscious, ers invest in seasonal products associated to festivities employed by street vendors to quickly recover their (e.g., Christmas and Halloween) and sports competitions. investments. Selling this type of merchandise potentially comes with We explore price discrimination in the market of stick- greater returns, but it is also tightly connected to higher ers for the FIFA World Cup’s album. The tradition of economic risks. Once the season is over, vendors have completing this collection has persisted over decades to drastically reduce prices to clear stocks or store them in Colombia, despite the arrival of digital information for extended periods, which implies additional costs. to keep track of competitors (teams and players) in the World Cup. Among the different seasonal products asso - ciated with the World Cup, such as jerseys and flags, we *Correspondence: cesar.mantilla@urosario.edu.co opted to use stickers since they provide two advantages. Department of Economics, Universidad del Rosario, Bogotá, Colombia First, we can test different discriminatory strategies Full list of author information is available at the end of the article © The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. 6 Page 2 of 24 P. Zamora et al. since we are able to manipulate the stickers’ salience complete the collection, she might charge a larger price individually. Second, the buyer can credibly signal a low for the same sticker when it belongs to a shorter list, as likelihood of future interactions because buyers mostly if she anticipates a buyer’s higher valuation for the last recur to street vendors when opportunities to trade with stickers left. other collectors have been exhausted, and the purchase The literature on market discrimination has mainly of sealed packets of stickers is likely to bring only dupli- focused on two types of discrimination. One of them cates. That is, buyers mostly interact with street vendors is taste-based discrimination, related to a preference- to obtain the last missing stickers in the collection. This driven mechanism (Becker 1957). The other is statisti - last feature allows us to minimize reputational concerns cal discrimination, a belief-driven mechanism relying on related to the intention of suppliers to offer lower prices self-fulfilling stereotypes (Arrow 1973). Whereas trans- to ensure future sales. actions data provide evidence of discrimination in mar- We conduct two studies to explore three potential kets associated with labor, health, housing, tourism, and sources of discrimination. In the first study, we evalu - retail sales (Apollo 2014; Balsa and McGuire 2001; Baran- ate two sources of discrimination that might be inferred zini et  al. 2008; Chen et  al. 2018; Grytten et  al. 2011; directly from the buyer’s appearance and accent. These Yip and Wong 2014), the controlled environment from are gender and nationality. We have four experimenters experiments has been useful for pointing out more spe- quoting a list of nine stickers. Two of them are Colom- cific discrimination mechanisms (Ayres and Siegelman bian and two of them are Argentinian, and there are a 1995; Auspurg et al. 2017; Baert and De Pauw 2014; Bal- woman and a man corresponding to each nationality. dini and Federici 2011; Castillo et al. 2013; Flage 2018). In the second study, we evaluate the third source of dis- Conducting an experiment involves an implicit trade- crimination. It relates to the quoted set of stickers (i.e., off between control and external validity. On the one the length of the stickers’ list). We have four Colombian hand, an experiment closer to the laboratory setting experimenters. Two of them quote a long list of nine provides greater control, making it easier to test spe- stickers, and two of them quote a short list of only four. cific mechanisms related to discrimination. On the other There are one woman and one man corresponding to hand, natural experiments emphasize realism while hold- each type of list. ing sufficient control to detect discrimination patterns Discrimination based on nationality is worth being (Charness and Kuhn 2011; List and Rasul 2011). Corre- explored in this context for two reasons. First, since spondence studies, within the family of natural experi- sports competitions increase national identity salience ments, aim to compare labor outcomes between testers (Turchin, 2016), it allows to test discrimination based (“confederates” or “auditors”) that, by assumption, differ on in-group favoritism and out-group hostility. Dis- only in the characteristic of interest for the analysis of crimination based on group boundaries has been tested discrimination, which is typically gender (Bertrand and in the context of a previous FIFA World Cup in a mar- Mullainathan 2004; Azmat and Petrongolo 2014). ket for low-cost electronic devices in Brazil (Kim et  al. The audit study technique employed in this paper is 2018). Second, foreigners might be subject to discrimi- methodologically close to correspondence studies, and nation based on the sellers’ belief that willingness to pay it is based on controlling the characteristics of the test- is higher for foreigners than locals. For developing coun- ers on one side of the market. When interacting with the tries with an increasing tourism industry, it is essential other side of the market, it aims to match most of the to understand whether informal markets, particularly attributes between testers except the one of interest for street-vending, are discriminatory against foreigners and detecting discrimination. In correspondence studies, the thus have reputational consequences for the whole tour- analysis focuses on the job-offer stage and may include ism industry. prices (List 2004) and financial or medical advice (Mul - We are also interested in gender discrimination given lainathan et  al. 2012; Currie et  al. 2011). In our audit, the prevalent stereotypes associating masculinity and there is a broader set of outcomes. In the experiment, sports competitions, particularly in Colombia (Csizma testers act as potential buyers and quote a set of prede- et  al. 1988; Grabow and Kühl 2019; Oxford and McLa- fined stickers, while street vendors provide prices in their chlan 2018). Sellers might assume that women have natural environment (remaining unaware of the audit). less experience buying stickers or are less knowledge- able about stickers’ prices and, therefore, discriminate accordingly. The third element useful to analyze price discrimi - Recent models bring other mechanisms to the discussion, including unin- nation is the length of the quoted list. Since the vendor tentional (or implicit) discrimination, search frictions in the labor market, and knows that the buyer is quoting the few stickers left to endogenous group formation (Bertrand et al. 2005; Fang and Moro 2011). Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 3 of 24 6 We contribute to this literature by conducting an is, two to three months in advance of the season’s high audit experiment in the context of an informal market, peak. where the lack of enforcement on mandatory posted In 2018, Completing the collection of 670 stickers cost prices increases sellers’ discretionary power (Grossman at least 36% of the monthly minimum wage in Colom- and Honig 2017). We find evidence of price discrimina - bia. Hence, collectors tend to be located in the city’s eco- tion based on the buyer’s nationality: vendors charge on nomic centers and high socio-economic residential areas. average 4.8% more per sticker to Argentinians, and the Vendors react accordingly and are located in areas with a effects are significant for the most expensive and the least large influx of office workers, forming clusters of vendors, expensive ones. Our conjecture for this finding is that and in the surroundings of stores of a high-end super- expensive stickers have greater price fluctuation, mak - market chain. ing it easier for vendors to hide their intention to quote Depending on their budget and time constraints, col- higher prices. Regarding the least expensive stickers, lectors recur to different strategies when buying sealed some vendors appear to believe that foreigners are less packets becomes too expensive due to the frequency of familiar with the low price charged for stickers printed duplicates. One alternative is exchanging stickers at parks in regular paper (instead of shiny foil). By contrast, we or malls, where gatherings coordinated via social media do not find evidence of price discrimination based on occur. However, they can be time-consuming when the buyers’ gender, and neither on the length of the quoted stock of duplicates is not very large. Another alternative sticker list. for collectors is to interact directly with the vendors. The We complement this study with a qualitative analysis advantage is that they tend to have a much larger set of based on eleven interviews with street vendors. We focus stickers to sell (and sometimes trade, although in much on describing their pricing strategies, their awareness of more favorable exchange terms for vendors). Therefore, price discrimination, and their perception (and trade) collectors whose budget constraints are looser prefer this of counterfeit stickers. The results from the interviews strategy. reveal that vendors do not seem to discriminate con- The price charged on the streets for stickers printed in sciously. Ten out of eleven interviewees state that they regular paper (players, pictures of teams, and stadiums) do not charge higher prices to foreigners, whereas about dwells between 1.25 and 2 times the sticker’s unit cost half of them admit selling counterfeit stickers. This result in a packet. By contrast, special stickers printed in shiny weakens the alternative argument that a social desirabil- foil tend to have a much larger price. For the 2018 World ity bias drives their responses to price discrimination. Cup, the fifty shiny foil emblems were scarce. Conse - quently, the price charged by sticker vendors reached 2 Experimental setting between 15 and 50 times the sticker’s unit cost in a 2.1 Description of the stick ers market packet. An alternative strategy was to buy the last miss- Street vendors, who represent a significant part of the ing stickers directly from the manufacturer. However, informal labor force in Bogotá, Colombia, are not mar- the stickers that could be purchased under this strategy ginalized and typically can afford the acquisition of sea - became unavailable in Colombia one month before the sonal products (Martínez et al. 2017). During the Football World Cup kick-off (three weeks before our data collec - World Cup, these products include jerseys and flags, and tion process). As in many other street-vending activities, stickers sold individually. The sale of individual stickers sticker vendors are used to bargaining prices with poten- is popular among street vendors because they are aware tial customers and have fast interactions that do not lead of the high costs of completing the collection and the to a transaction in a considerable number of cases. associated willingness to pay for individual stickers from a considerable number of collectors. Since vendors have high expectations about the World Cup season, they aim to accrue several duplicates of each sticker beforehand to While the collection only includes 670 stickers, the stock of some vendors reached about 10,000 stickers. respond to future demand. Vendors usually start open- A collector would need to buy 134 packets containing five stickers each, ing boxes as soon as the producer releases stickers. That for a total price of 281,400 COP. The monthly minimum wage for 2018 was 781,242 COP. This amount is a lower bound for the cost, under the implau- sible assumption that the collector would trade any duplicate she obtains. At the time of the experiment, $1 US Dollar was equivalent to (approxi- mately) 2,900 COP. 2 5 To give an idea on the high valuation of stickers, the 2018 census on inven- A Colombian newspaper issued an article entitled “Panini will no longer tory reduction and theft prevention (Censo Nacional de Mermas y Prevención ship stickers for the Russia’s World Cup album” (https ://perma .cc/X34B- de Pérdidas) revealed that sticker packets were the fifth most stolen item from DHGE). Unlike in the European Union and the United States, collectors supermarket chains in the whole year (https ://perma .cc/KGM4-MHAH), cannot directly purchase the complete sticker collection from the official after canned food products, soft-drinks, candies, and toiletries. distributor’s website. 6 Page 4 of 24 P. Zamora et al. 2.2 F ield site and vendors’ selection We avoided buyers’ overlap in the same cluster by delay- We exploit the seasonal demand for album stickers of ing the data collection from one of the experimenters. the Football World Cup 2018. Six experimenters acted Table 6 (see Appendix 1) details the list of days in which as potential buyers and approached fifty-nine street each experimenter quoted the prices. vendors. We define as vendor any subject, or group of subjects, who sell stickers in a fixed or mobile stall. 2.3 Execution of the audits Regardless of the type of stall, vendors’ locations are Members from the research team collected the quoted fixed. Hence, two or more subjects who share a stall are prices in pairs. One of the team members who were ini- treated as a unique vendor. In consequence, it is possi- tially in charge of mapping vendors’ locations always ble that two different buyers obtained the prices from the accompanied the buyer. In this way, the companion team same vendor ID but interacted with different subjects. member could verify that the buyer was approaching The vendors included in our sample were located in the right vendor. Moreover, we only made contact with five different clusters along the Eastern side of the city, the vendor when there were no potential buyers around. between Street 13th (downtown) and Street 147th. These The buyer approached the vendor and asked for a list of clusters coincide with centers of economic activity or res- stickers’ prices, while the companion waited at a prudent idential areas of medium–high and high socio-economic distance. The opening line of the script, leading to the status. Other areas of the city were not sampled because vendor–buyer interaction, says: sticker vendors were scarce and sparsely distributed. The “My husband/wife and I are looking for the last week before the data collection, two members of the stickers to complete our kid’s PANINI album. He/She research team mapped the locations of seventy-six street is also looking for the stickers right now, so for the vendors across the five geographic clusters and took pic - moment, I am only quoting. Do you complete lists (of tures of their stalls. missing stickers)?” Based on the initial mapping of vendors, we defined as an inclusion criterion that neighboring vendors must be Given the nature of street-vending in Colombia, quot- at least two meters away from each other (i.e., we chose ing prices without committing to a purchase is an every- them “randomly,” conditional on being at least at a 2  m day interaction. Nonetheless, it is important to discuss distance). We excluded from the sample eighteen ven- two trade-offs created by this script. First, it increases the dors that were too close to each other. Our final sample internal validity. However, it comes at the expense of a includes fifty-nine (78%) of the initially mapped vendors. greater likelihood that sellers could detect that they were Figure  2 shows a detailed description of the vendors’ part of an audit. Second, quoting rather than purchasing location (see Appendix 1). We have substantial variation stickers is useful to leave unaltered the stock of specific in the inclusion rates across clusters. Whereas in the two stickers. Nevertheless, neighboring vendors observing a south-most clusters, the inclusion rate goes above 91%, in failed transaction might update their perception of the the other three clusters, this rate dwells between 53 and buyer’s bargaining skills. We argue that the two caveats of 72%. The inclusion rate is particularly low in the cluster using the script are dampened by the exclusion rule used of Street 147th, where sticker vendors rent temporary to sample vendors and by leaving sufficient time between tents located very close to each other (there are fourteen interactions from the same vendor with two different stalls in 120 linear meters). In our two studies, the inter- research team members. actions with street vendors took place between (Tuesday) Mentioning that the stickers are for the buyer’s kid, June 5th and (Friday) June 8th, one week before the kick- and that these are being simultaneously quoted for both off of the 2018 World Cup. We conducted the experi - parents, helps reducing the commitment for a purchase. ment within four days, on the same week, to minimize Moreover, mentioning the kid on the script aims to pre- substantial variations in the vendors’ stock of stickers. At vent vendors from inferring that men were quoting the most, two buyers were on the field quoting prices on any stickers for themselves, whereas women were doing this particular day. They started to quote lists as early as 9:00 task for someone else. We aim at blocking the latter pos- a.m. and finished as late as 7:30 p.m. One of them visited sibility, which might be interpreted as a confounder of the vendor clusters from South to North of Bogotá, and gender-based discrimination with what  Balafoutas et  al. the other buyer did this process in the opposite direction. (2015) define as “second-order moral hazard.” That is, the tendency of the supply side (i.e., sticker vendors) to anticipate the moral hazard on the demand side (i.e., One week before conducting the experiment, two research team members randomly asked some vendors their usual schedule in each of the sample zones. Given this information, we could guarantee that most vendors would be present when experimenters quoted prices the following week. See Appendix 2 to read the full script. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 5 of 24 6 women quoting stickers for a third party), which results First, 60% of our sample of vendors are close enough to in changes in the price associated with the inference of each other, and therefore looking for these stickers has whom is the final payer. In our design, mentioning the a low transaction cost. Second, vendors typically do not kid makes more likely that male and female buyers are get involved in monetary transactions between them. assumed as final payers of the stickers. Instead, the vendor borrows stickers from another ven- Buyers dressed similarly, with semi-formal clothes to dor to secure the transaction with the buyer, and then represent that they were in their late twenties or mid- gives back the same sticker, or an equivalent one to the thirties, close to their actual age, ranging from 24 to lender. A broad definition of equivalence is defined by the 40 years old. Once the vendor confirms that she/he sells four categories mentioned before: front page emblems, individual stickers, we start reading aloud a handwritten “Legends,” team emblems, and regular stickers. These list of stickers (the set included on the lists is explained “equivalences” also make more plausible the assumption in the next section). The buyer registered every quoted that “mental search costs,” defined as the cognitive costs price of the list, element by element. In case that the ven- of recalling the price of stickers previously sold, are not dor did not have the sticker, we asked for the price she very high either. would charge to find it. In case the vendor said that she The buyer carried a handwritten list and a printed could not find the sticker, we asked for the price she had form. Both included the list of stickers. The latter added previously charged for it in past transactions. This pro - information buyers had to collect prior to the interaction: cedure allowed us to increase the chances of obtaining vendor’s gender, an assessment of her/his age, whether information from every quoted sticker, even if it was not she/he sells other products (related or not to the World available at the time. Cup), type of stall, date and time of the interaction (plus One might think that this procedure makes the audit some empty lines for taking relevant notes regarding the seem “artificial,” as if the buyer put more emphasis on interaction). learning about the prices rather than looking for a transac- Buyers used the handwritten list during the interaction, tion. We provide two reasons why we consider this is not a so if the vendor wanted to grab the list, it would not have major concern. First, from the beginning of the interaction, revealed any additional information about the experi- the buyers mentioned that they were, for the moment, ment. Different members of the research team made the only quoting the stickers. Consequently, since any transac- handwritten lists. After finishing the interaction with the tion would occur another day, asking for the price of cur- vendor, the buyers walked away, hid from the vendor’s rently unavailable stickers would make sense if they could sight, and filled out the printed form. be obtained on a future date. Second, although there are 670 stickers, the total number of price categories is much 2.4 Selection of stickers lower (see Sect. 2.4). Therefore, asking about the price of The FIFA World Cup album’s stickers can be divided into a currently unavailable sticker would have been equiva- two main groups: those printed in shiny foil and those lent to quote another sticker from the same price category printed in regular paper. The latter corresponds to 92.5% (and the latter type of question would have seemed more (620/670) of the entire collection and includes photo- artificial). graphs from players, teams, and stadiums. Most collec- An additional concern about this procedure, in which tors exchange on a 1-to-1 basis these “regular” stickers, we quote unavailable stickers, is that  prices might and most vendors (69%) charge around 1.0 and 2.5 times include search costs. Although this is a valid concern, we the unit cost (i.e., the cost of the packet divided by five, observe that the likelihood that vendors did not have the the number of stickers in the packet) for them. Tables  1 requested sticker but mentioned a price for getting it was and 2 display the list of selected stickers for each study. low. Table  7 (see Appendix 1) reveals that, on average, it We included three “regular” stickers (a stadium, a team, only occurred in 4.35% of the vendor–buyer interactions and a player) in the long lists. The purpose of including (and for 2% of all the potential quoted stickers). Moreo- these stickers was twofold: (i) make more salient the dif- ver, we believe that search costs were low for two reasons. ference between the long list (9 stickers) and the short list (4 stickers) in Study 2, without considerably affecting the difference in total prices, and (ii) make more engaging and realistic the long list for sticker vendors. Vendors recognize four categories of stickers: front page emblems, “Leg- The fifty shiny foil stickers have three main categories ends,” team emblems, and regular stickers (stadiums, teams’ picture, and play- and are significantly more expensive than regular stick - ers). ers. The eight stickers on the front page of the album are The likelihood that vendors did not have the requested sticker and did not mention a price for getting it was also low. Table  7 shows that, on average, the first category. These stickers are related to the World it only occurred in 4.17% of the vendor–buyer interactions (and for 2.1% of Cup (e.g., logo, poster, mascot, ball). The second category the total number of quoted stickers). 6 Page 6 of 24 P. Zamora et al. Table 1 Study 1: Composition of stickers’ lists and variation in buyers’ attributes Experimenter Information ID 1 2 3 4 Nationality Argentinian Argentinian Colombian Colombian Gender Female Male Female Male List quoted Long Long Long Long Sticker type Stickers quoted by each experimenter Local team emblem Colombia Colombia Colombia Colombia Front page emblem 1 FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy Front page emblem 2 Ball Telstar Ball Telstar Ball Telstar Ball Telstar World Champion sticker Maradona (ARG) Maradona (ARG) Maradona (ARG) Maradona (ARG) Team emblem tier 1 Argentina Argentina Argentina Argentina Team emblem tier 2 Uruguay Mexico Peru Uruguay Stadium Samara St. Petersburg Spartak Nizhny Novgorod Team Iceland Denmark Denmark Iceland Player Busquets (ESP) Lallana (ENG) Lallana (ENG) Busquets (ESP) Total Number of stickers quoted 9 9 9 9 on the list Every experimenter quoted one sticker from each row. Most of the stickers are printed on shiny foil, except the stadiums, teams, and players, printed on regular paper. Although we cannot have a formal test to check that the stickers quoted in each category by different buyers are the same, we know that sellers, in general, assign prices based on a limited number of sticker types. Vendors usually do this because having prices for each of the 670 stickers is an enormous cognitive cost for them, and it also increases the menu costs. Additionally, by the very nature of street sales, these are fast transactions, and sellers usually mention prices quickly when one asks for a quotation. Table 2 Study 2: Composition of stickers’ lists and variation in list’s length Experimenter Information ID 3 4 5 6 Nationality Colombian Colombian Colombian Colombian Gender Female Male Female Male List quoted Long Long Short Short Sticker type Stickers quoted by each experimenter Local Team Emblem Colombia Colombia Colombia Colombia Front page Emblem 1 FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy Front page Emblem 2 Ball Telstar Ball Telstar Panini Logo Fair Play Logo Legend Maradona (ARG) Maradona (ARG) x x Team Emblem Tier 1 Argentina Argentina Brazil Germany Team Emblem Tier 2 Peru Uruguay x x Stadium Spartak Nizhny Novgorod x x Team Denmark Iceland x x Player Lallana (ENG) Busquets (ESP) x x Total Number of stickers quoted 9 9 4 4 on the list Every experimenter quoted one sticker from each row. The "x" denotes that this sticker was not quoted in the Short List treatment. The quoted stickers are printed on shiny foil, except the stadiums, teams, and players, which are printed on regular paper comprises thirty-two team emblems, one per each par- the quoted prices that the team emblems are the least ticipating team. Finally, ten stickers presented on the last expensive among these three categories. An exception page of the album form the last category. These stick - is Colombia’s team emblem. Given that the audit was ers are called “Legends” and correspond to the existing conducted in Bogotá, vendors tend to charge the high- World Champions and top-scorers. We learned from est price for this sticker. Since the shiny foil stickers are Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 7 of 24 6 more expensive than the regular ones, in particular those of the audited vendors, allowing us to introduce seller’s on the front page, plus Colombia’s emblem, and the “Leg- fixed effects into the econometric specification. With end,” they open a broader space for price discrimination, this caution in mind, we can check gender differences by either gender-based or nationality-based. comparing prices given to the experimenters with IDs 1 The remaining six stickers included in Study 1 are and 3 vs. 2 and 4; and we can check differences by nation - printed in shiny foil. Two of them are stickers from the alities by comparing prices given to experimenters with front page, one is a “Legend”, and the other three are team IDs 1 and 2 vs. 3 and 4 (see Table 1). emblems. Given the buyers’ nationality, we decided to incorporate Colombia’s emblem and Argentina’s emblem 2.6 Study 2: Price discrimination based on sticker list’s (team in Tier 1). The third emblem is a Latin American attributes team in Tier 2 (either Mexico, Peru, or Uruguay), which The purpose of collectors of the FIFA World Cup album serves as a baseline cost for team emblems. is to complete the whole set of stickers. We argue that, In Study 2, our primary purpose is to detect price dif- when there are few stickers left to complete the collec- ferences per sticker when they belong to a long or a tion, it is plausible that each additional sticker’s marginal short list. In the audit experiment, the long list is identi- value is increasing. In other words, the sticker is valuable cal to the one from Study 1 (indeed, observations from per se, and because collectors derive utility from getting two buyers overlap between studies). For the short list, closer to the completion of the set. Thus, we can test experimenters quoted two stickers from the front page, whether street vendors adopt a sophisticated pattern of Colombia’s emblem, and a Tier 1 emblem: Brazil or Ger- price discrimination, in which they extract a signal of the many. In this sense, the short list included three stickers buyers’ increasing valuation of the missing stickers based where price discrimination might be more pronounced, on the length of the list. We conducted a second experi- plus the Tier 1 team’s emblem as a baseline within the ment in which we hold constant the buyers’ nationality short list. (i.e., all of them are Colombian) to focus on the length of the quoted list. We devised a short list of stickers, com- parable to a subset of what we refer to as the long list 2.5 S tudy 1: Price discrimination based on buyers’ employed in Study 1. attributes Table  2 compares the long list for experimenters with Four experimenters were acting as (potential) buyers. IDs 3 and 4 (i.e., the same data collected in Study 1) with They differed in nationality (two Colombians and two the short list for the new experimenters with IDs 5 and Argentinians) and gender (two males and two females). 6. To test whether the vendors charge more per sticker We expected that vendors could infer that foreigners when the inferred marginal valuation of stickers is higher, were Argentinians due to their pronounced accent; and we compare the quoted prices on the Long List treatment due to some slight variations in the script. For example, (IDs 3 and 4) to the quoted prices on the Short List treat- the script for the Argentinian buyers included the under- ment (IDs 5 and 6). lined words: I am missing one legend, number 664, the one with 3 Data the best player in history: Maradona. The guy is a We collected information on interactions between 59 genius! How much is it? street vendors and 6 experimenters or buyers. We thus Although these additional words make the script have a total of (59 × 6) 354 potential vendor–buyer inter- slightly different for Argentinians, we argue that the actions. We gathered 287 (81%) of such interactions, 199 benefit (i.e., remarking the foreigner buyer’s nationality) for the Long List, in which we had four buyers; and 88 for exceeds the cost (i.e., the risk that these words, and not the Short List, in which we had two buyers. The missing nationality per se, drive the price differences). interactions correspond to cases where the experimenter We exploit those differences to detect price variations could not find a specific vendor. Figure  3 in Appendix 1 between nationalities and between genders. Table  1 maps the interactions of each buyer with vendors. The reports the list of stickers that experimenters quoted in average number of interactions per vendor was 4.86, Study 1. We present the nine quoted stickers in descend- and the median was 5. Forty-one percent of the vendors ing order according to their average quoted price. interacted with all the experimenters, whereas fourteen We acknowledge the limits of our study’s “experimen- percent of the vendors interacted with less than four tal” character since neither gender nor nationality of the experimenters. Taking “sticker × interaction” as the unit buyers were randomly assigned (Siegelman and Heck- of observation, we have a total of 1712 observations in the man 1993). Nonetheless, our design aims to overcome Long List treatment corresponding to four experiment- this limitation by having multiple interactions for each ers and 302 in the Short List treatment corresponding 6 Page 8 of 24 P. Zamora et al. to two experimenters. Note that 1712 is lower than the tend to have more interactions during the afternoon (i.e., (199 × 9 =) 1791 potential observations in the Long List, after 12:00  m) compared to female vendors. We control and 302 is lower than the (88 × 4 =) 352 potential obser- for these variables in our regression analysis. vations in the Short List. The reason is that some vendors did not have some stickers or refused to quote the prices 4 Results individually and instead provided the price of the entire 4.1 Study 1: Econometric results for discrimination based list. on buyer’s attributes Regarding vendors’ characteristics, we find that 42% We estimate a simple linear regression model with robust were in their stall without any companion. According to standard errors and vendor fixed effects to test whether our measure of perceived age, 61% of sellers were 35 years street vendors quote different prices based on the buyer’s old or younger. Concerning vendors’ location, we find nationality or gender. The estimated model is given by: that 43% of vendors are far from any other vendor (more ′ ′ Price = β Foreign + β Male + X α + T δ + ǫ sij 1 j 2 j sij i sij than 50  m away), 44% are at a prudent distance from at (1) least another vendor (between 10 and 50  m away), and the remaining 14% are all located very close to each other The dependent variable is the price of sticker s quoted by (between 2 and 10  m away). As mentioned before, we vendor i to buyer j, expressed in thousands of Colombian excluded from the sample any vendor at a closer distance. pesos [kCOP]. The indicator Foreign equals one if the buyer The right panel of Figure  2 in Appendix 1 shows each is Argentinian and equals zero if the buyer is Colombian. vendors’ location within their geographic cluster. Ven- Similarly, Male equals one if the buyer is a man and equals dors were particularly close to each other in the clusters zero otherwise. In this equation, X is a vector of vendors’ located in Street 85th and Street 147th. For the statisti- characteristics, including age, gender, geographic location cal analysis, we defined only two categories of distance to (cluster), and the presence of neighboring sticker vendors. other vendors: being far from (i.e., more than 50 m away) The vector T includes sticker fixed effects, a categorical isj or close to (i.e., less than 50 m away) any other vendor. variable for the day of the week in which the interactions Since most buyers quoted the price of stickers with took place, and a dummy variable for whether the interac- every vendor, we analyze the interactions as a within- tions occurred in the morning or in the afternoon. subjects design. Nonetheless, we check whether vendors’ Results from Table  3 reveal that prices per sticker characteristics are balanced across the experimenters’ quoted to Argentinians are 208 COP higher than those features of interest in Study 1, and across types of lists quoted to Colombians (about 4.8% higher with respect and experimenters’ features in Study 2, for two reasons. to the mean price of the 1712 stickers in the regression). First, we have some degree of attrition because the exper- However, we do not observe a systematic difference in imenters could not find all vendors. Second, according to prices between male and female buyers. Although male our assignment of vendor ID, the experimenters might buyers are quoted 290 COP more per sticker than female have interacted with different people located in the same buyers, standard errors are slightly larger than the coef- stall. ficients, and therefore we cannot reject the absence of Table 8 in Appendix 1 shows the coefficients of four lin - price discrimination based on the buyer’s gender. Models ear probability models in which we take as covariates the 1 and 2 show that the results are similar using classical or vendors’ gender and (perceived) age, whether the vendor robust standard errors. Given the small number of exper- was isolated, and the journey (morning or afternoon). imenters acting as buyers, we cannot cluster the standard None of the covariates are correlated with the buyer’s errors at the “pseudo-randomization” level. nationality. However, in Study 1, vendors perceived as The econometric model includes seller fixed effects. This older are more likely to interact with male (instead of specification allows us to exploit the within-subject nature female) buyers. In Study 2, male vendors in the Short List of our sampling strategy. In other words, the observed dif- ferences in quoted prices based on buyers’ attributes come from (a subset of) sellers quoting different prices to local and foreign buyers. In this case, the difference we obtain is Three vendors out of forty-eight gave the total price for the list of nine more robust than what would have been obtained by com- stickers to the Colombian buyer instead of quoting the prices individually. Seven vendors out of forty gave the total price to the male buyer for the list of paring quoted prices given by sellers that interacted with four stickers and did not mention individual prices. In the case of the female local buyers to prices given by (comparable) sellers that inter- buyer quoting the short list, five out of forty-eight vendors preferred to give acted with foreign buyers (between-subjects design). Fig- the total price for the four stickers instead of mentioning prices individually. ure 1 depicts some coefficients of interest from Eq.  1. Panel The statistics presented in the text correspond to the average for the six experimenters. Since it is possible that buyers interacted with different A shows that vendors quote a higher price for shiny foil stick- vendors at the same stall, we first compute these demographic statistics per ers, particularly for stickers on the album’s front page and experimenter and then we average the reported percentages. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 9 of 24 6 Table 3 Eec ff t of buyer’s attributes (nationality and gender) on One of the advantages of creating a list with multiple street vendors’ price quotations items to quote is the opportunity to explore more sophis- ticated patterns of price discrimination. Since we verified Variables (1) Price [kCOP] (2) Price [kCOP] the existence of price variation across sticker types (see Foreigner 0.208* (0.113) 0.208* (0.120) Panel A in Fig. 1), we now conduct the regression analysis Male Buyer 0.290 (0.303) 0.290 (0.369) separately for each sticker on the list. Constant − 0.168 (0.326) − 0.168 (0.284) Table 4 reveals the results for the price quoted for each R-squared 0.799 0.799 sticker. Additionally, Column 1 shows the price quoted Controls and Yes Yes for the full list. This result confirms that vendors charge Vendor FE larger prices to Argentinians than to Colombians (on Errors Standard Robust average, 2,438 COP more). Nonetheless, this pattern Vendors 59 59 is not uniform across stickers: the second most expen- Number of 1712 1712 sive sticker (i.e., the FIFA Trophy), and the two cheapest observa- stickers on the list, drive the higher prices quoted to for- tions eigners. The coefficients for the two other most expensive *p < 0.1, **p < 0.05, ***p < 0.01. Standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in stickers, Colombia’s emblem and the Official ball-Telstar, thousands of Colombian pesos (kCOP). The mean of the dependent variable show a pattern similar to the FIFA Trophy, though they is 4.275 kCOP. Controls in all regressions: geographic cluster, day and sticker are less precisely estimated. fixed effects; (perceived) age and gender of the vendor, whether the vendor has nearby vendors, and time of the day We offer a conjecture for this pattern, in which Argen - a Stickers Colombia's FIFA World Ball Maradona Argentina's Tier 2 Team Stadium Team Player Emblem Cup Trophy Telstar (Legend - ARG) Emblem Emblem b Zones Downtown Street 72 Street 85 Street 93 Street 147 Fig. 1 Coefficients for additional covariates in the regression shown in Table 3: (A) sticker fixed effects, (B) geographic clusters. The color intensity of confidence intervals represents the confidence level (from darker to lighter, 90, 95, and 99%). The reference category in Panel A is the sticker of a player (i.e., the cheapest sticker). The reference category in Panel B is downtown (the area of Bogotá in which stickers are the cheapest) Colombia’s emblem. Panel B reveals that stickers’ price dif- tinians are simultaneously charged more for the most and fers between spatial clusters, increasing as we move to the the least expensive stickers. On the one hand, vendors Northern side of Bogotá. have more discretionary power to raise prices of the most expensive stickers, arguing that they are in high demand. An explanation for the considerably larger price in the farthest cluster One could interpret this result based on the heteroge- (Street 147th) is that sticker vendors rent temporary tents located in a small mall’s borders at a significant price (about 1.5 monthly minimum wages). It neity in price discrimination detection (Grossman and leads to a very high concentration of vendors (fourteen in 120 linear meters) Honig 2017). This pattern is not observed for other shiny that, despite their spatial closeness, manage to keep the city’ highest prices. foil stickers with a small variation in their price, such as However, given their spatial proximity and their long-lasting reputation of selling stickers in every World Cup, it is also possible that vendors in this area team emblems. On the other hand, for teams and players, are more likely to collude. kCOP kCOP 6 Page 10 of 24 P. Zamora et al. ff Table 4 Eect of buyer’s attributes (nationality and gender) on quoted price, per type of sticker (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Full List of nine Colombia’s FIFA World Cup Official ball Maradona Argentina’s Emblem Tier 2 Team Team Player Variables Stadium Stickers Emblem Trophy -Telstar (Legend - ARG) Emblem Foreigner 2.438* (1.305) 0.467 (0.338) 0.680** (0.300) 0.328 (0.284) 0.042 (0.328) − 0.010 (0.191) -0.063 (0.208) 0.108 (0.099) 0.174* (0.092) 0.161* (0.089) Male Buyer 3.200 (6.334) 1.133 (1.450) 0.564 (1.054) 0.192 (1.111) − 0.273 (0.967) − 0.074 (0.655) 0.499 (1.029) − 0.050 (0.233) 0.087 (0.206) 0.224* (0.123) Constant 33.941*** (4.856) 3.920*** (0.969) 6.254*** (1.121) 5.968*** (1.170) 6.225*** (0.883) 4.442*** (0.589) 4.235*** (0.685) 0.747*** (0.212) 0.750*** (0.189) 0.285*** (0.099) R− squared 0.766 0.817 0.749 0.716 0.668 0.699 0.596 0.736 0.715 0.522 Controls Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes and Ven- dor FE Vendors 59 59 59 59 59 59 59 59 59 59 Number of 187 193 187 187 188 192 192 190 191 192 observa- tions *p < 0.1, **p < 0.05, ***p < 0.01. Robust standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in thousands of Colombian pesos (kCOP). Controls in all regressions: geographic cluster, day and sticker fixed effects; (perceived) age and gender of the vendor, whether the vendor has nearby vendors, and journey. In Column (1), we use the list’s total price if the vendor mentioned it. Otherwise, we use the sum of the nine stickers for each vendor–buyer interaction when the vendor separately quoted the prices for the nine stickers. We could not construct the sum in 12 interactions (10 vendors) because the vendor did not quote some stickers. In 9 interactions (8 vendors), the vendors gave a total price for the nine stickers. Four of them did not say the price for each sticker individually, and five of them proposed to give the Stadium, the Team, and the Player for free if the buyer bought the other six stickers. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 11 of 24 6 the least expensive stickers on the list, the modal prices for each of the listed (and comparable) stickers. We do are 1000 and 500 COP, respectively. That is, they are not find evidence suggesting that vendors charge a larger between five and ten times cheaper than the least expen - price for the same sticker when it belongs to a short list sive shiny foil stickers. Figure 4 reveals that foreigners are compared to when it belongs to a long list. Although the more likely to be charged prices two or three times larger number of observations is smaller in this econometric than the modal price for these regular stickers. Another exercise (the sticker list is shorter and Argentinians were mechanism that might be at play for the least expensive excluded), coefficients for the indicator variable for the stickers emerges from the fact that regular stickers are Long List variable are very close to zero. In other words, sometimes offered “as a gift” when purchasing the shiny the lack of significance does not reflect a lack of statistical foil stickers. However, we do not find evidence that these power. gifts are more likely to be offered to Colombians. We have two conjectures for the absence of an effect. Regarding the analysis of gender differences, Table  4 The first one is that sellers do not exploit the (assumed) provides additional, suggestive evidence that male buy- higher marginal valuation for fewer missing stickers. ers do not appear to receive systematically different The second conjecture is that, although one of the lists is quoted prices compared to female buyers. Nonetheless, twice as long as the other, both lists could be considered these results must be taken with caution since the sam- by the vendor to be “short,” in absolute terms. ple size sharply decreases at the sticker level. Similar to the discrimination against Argentinians, it might be the case that men are also charged higher amounts for the 5 Qualitative results from interviews most expensive stickers, particularly for the Colombian In addition to our audit studies, we conducted eleven emblem. interviews with sticker vendors that were part of our As an additional check, we add to the econometric original sample. At least two vendors were inter- model the interaction between the buyer’s and the ven- viewed per geographic cluster. The purpose was to dor’s gender. Table 9 reveals that gender interaction vari- complement our insights from this market and elicit ables are not statistically significant. Therefore, even if sellers’ perceptions of price discrimination. The we consider more specific mechanisms in which male interviews were conducted four weeks after the audit vendors might discriminate against male buyers but not experiment. By that time, the knockout-stage of the against female buyers or vice versa, we do not detect FIFA World Cup was already taking place, and the sale price differences based on gender. of stickers was less intense, so it was feasible to ask sellers to participate in our interview. The interviews were conducted on July 2, 2018, an official holiday in 4.2 S tudy 2: Econometric results for discrimination based Colombia, for the two clusters corresponding to resi- on list’s attributes dential areas. We selected this day because there were We discussed earlier that stickers might have an increas- substantially fewer people, and the probability of the ing marginal utility when the buyer’s purpose is to vendor accepting the interview was higher than on a complete the stickers collection. Vendors could then business day. The other three clusters correspond to anticipate a greater willingness to pay for the same sticker economic activity centers, and sticker vendors do not from a buyer revealing a short list of missing stickers, work on holidays. Hence, we conducted the interviews with respect to a buyer revealing a long list of miss- the day after. ing stickers. We conduct a regression similar to Eq.  (1), We opted for conducting eleven interviews due to replacing the variable that captures buyer’s nationality budget constraints since we offered compensation for (since all the experimenters in Study 2 are Colombians) the vendors’ time. We selected the vendors based on for a variable indicating whether the quotation corre- three attributes. First, vendors for whom we had com- sponded to a short or a long list. We focus here on the pleted the interactions with each experimenter (six in quoted price for the four stickers listed in Table  2 that total). The reason was to maximize the chances of find - are common to both lists: Colombia’s emblem, the FIFA ing the selected vendor by the time of the interview. Sec- World Cup Trophy, an additional shiny foil sticker from ond, since we had six observations per selected vendor, the front page (i.e., the Official ball-Telstar, the FIFA Fair we aimed at guaranteeing that it was the same person in Play logo or the Panini Logo) and the emblem from a Tier 1 team (Argentina, Brazil or Germany). We report the comparison between the Short List and One final interview was completed ten days later (July 13). We decided the Long List treatments in Table 5. The first column cor - to make this last interview to obtain an additional testimonial from the area responds to the model pooling all four stickers, whereas where vendors were closer to each other but from one of the few isolated ven- the next four columns display the coefficients of interest dors in this area. 6 Page 12 of 24 P. Zamora et al. Table 5 Eec ff t of sticker list’s attribute (length) on quoted price, per type of sticker Variables (1) Full list of four (2) Colombia’s Emblem (3) Front Page Emblem (4) Front Page Emblem (5) Tier 1 Team stickers 1 2 Emblem Long List 0.056 (0.113) 0.095 (0.237) 0.021 (0.150) 0.046 (0.155) 0.014 (0.127) Male Experimenter -0.029 (0.122) 0.221 (0.247) -0.204 (0.147) -0.261* (0.150) 0.013 (0.126) Constant 4.064*** (0.438) 3.034*** (0.714) 4.433*** (0.458) 4.322*** (0.424) 3.650*** (0.431) R-squared 0.553 0.803 0.838 0.811 0.709 Controls and Vendor FE Yes Yes Yes Yes Yes Number of observa- 697 176 172 173 176 tions *p < 0.1, **p < 0.05, ***p < 0.01. Robust standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in thousands of Colombian pesos (kCOP). Controls in all regressions: geographic cluster, day and sticker fixed effects; (perceived) age and gender of the vendor, whether the vendor has nearby vendors, and journey. Front Page Emblem 1: FIFA World Cup Trophy. Front Page Emblem 2: Official ball—Telstar, FIFA Fair Play Logo, or Panini Logo. Tier 1 Team Emblem: Argentina, Brazil, or Germany. In Column (1), we use the list’s total price if the vendor mentioned it. Otherwise, we use the sum of the four stickers for each vendor–buyer interaction when the vendor separately quoted the prices for the four stickers. We could not construct the sum in 3 interactions (3 vendors) because the vendor did not quote one of the four stickers. In 24 interactions (21 vendors), the vendors gave a total price for the four stickers. Eleven of them did not say the price for each sticker individually. all the interactions. Third, we guarantee one man and one respectively. The remaining one did not answer. Three woman were interviewed in each cluster. vendors declared to contribute to the pension system or The interviewer approached the vendor and asked if health system. she/he could respond to some non-personal questions Two interviewees in Street 147th were located in that would take between 15 and 20 min. We list the ques- fixed stalls. One paid a monthly rent of approximately tions in Appendix 1. 1,200,000 COP, and the other paid 1,700,000 COP (between 1.5 and 2 monthly minimum wages). The rest of the interviewees had a mobile stall. Although they could A. The interviewer would buy one of the shiny foil stick - change locations within or between days, sticker vendors ers for 10,000 COP in appreciation. By the time of the tend to stay in a fixed spot during the World Cup. All but interview, the price of such stickers was about 4,000 one of the interviewed vendors has stayed in the same to 6,000 COP. All the vendors initially approached spot since at least the 2010 FIFA World Cup. Despite by the interviewer agreed to participate. We did not the Constitutional protection of street vendors, they are mention any relationship between the interview and sometimes subject to police inspections that might lead the audit study, as it may have caused unpleasantness to confiscation of their merchandise. In these cases, ven - among the interviewees. We requested permission to dors with mobile stalls locate nearby their usual spot. record the interview, and all of them accepted. 5.1 Pricing strategies We interviewed five women and six men with the We identify three different types of strategies based on following characteristics. The interviewer perceived the responses from eight interviewees, which we define eight of them as being older than 35  years. Five of them as frequency-based, distributor-based, and imitation. declared to work in their stall alone, whereas the other The most usual strategy is what we define as frequency- six declared to work with at least one more person. In based. It consists of opening a vast number of boxes, each most of the cases, this companion person was not a rela- containing 104 packets with five stickers each, for a total tive. Working hours from interviewed vendors went from of 520 stickers per box, and then inferring each sticker’s 8 to 12.5 h a day, with an average of 10.4. Depending on rarity to assign prices based on how scarce they are. The the geographic cluster, they work on weekdays (Down- following is an example of a vendor using this strategy: town and 72nd Street) or the whole week (85th, 93rd, and 147th Street). Some of them claimed that they were [Excerpt 1] “Well, I already have experience with working more hours than usual given the World Cup the World Cup, right? I open boxes because you must Season. Eight of the interviewees were also dedicated open boxes or master boxes as people say… because to street vending the rest of the year, and they sell other seasonal products (e.g., Christmas merchandise) or spe- cific products such as flowers, fast food, umbrellas, and The responses of the three remaining vendors cannot be linked to any strat - used books. Another two vendors who replied to this egy. These were: “based on the price of the box”, “shiny foil stickers cost 4,000 COP, except #1 that is more costly, and any other sticker costs 300 COP”, and question worked in tailoring and building maintenance, “this is my fourth World Cup selling stickers, so based on my experience.” Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 13 of 24 6 a master box contains 12 boxes. So let’s say I open 3, average prices are also higher than in the other clusters. It 4, master boxes. Then, by arranging the stickers by is also worth mentioning that, whereas they perceive the numbers, I deduce the team that comes out the most, information provided by other vendors as useful, several the players who come out the most, and the stickers vendors also claimed that they dislike when buyers tell that come out the least. Hence, those that come out them that a nearby vendor is selling the same sticker for a the most are the cheapest, and those that come out lower price. Regardless of whether the prices provided by the least are the most expensive.” buyers were truthful or were part of a bargaining strategy, vendors appeared to dismiss this information. When asked about the number of opened boxes, the responses ranged from 24 to 100. This process gives 5.2 Awareness of price discrimination between 12,500 and 52,000 stickers, which will be suf- We included four questions regarding perceived price ficient to complete between 18 and 77 collections in the variations. First, we asked vendors whether they changed absence of duplicate stickers. The frequency-based strat - the prices of stickers within a given day. Ten of them egy requires a substantial investment since each box ini- responded negatively. The only vendor who provided tially cost them around 180,000 COP. Opening 24 boxes a different response claimed that he would change the would cost 4,320,000 COP, more than five times the price of stickers within a day if the number of units left monthly minimum wage at the time. of a specific sticker was too low. This vendor followed a The distributor-based strategy consists of following distributor-based pricing strategy. the prices per sticker suggested by their PANINI dis- We also asked vendors whether they had changed the tributor for most of the stickers. In general, vendors were prices of stickers between days. Ten of them responded instructed to charge 300 COP per sticker and comply affirmatively, arguing that prices had been declining over with most stickers. However, they tended to charge more time since mid-March when the distribution of stick- for the shiny foil stickers and others that they considered ers and albums started in Bogotá (the remaining vendor rare. responded negatively). Even though most of them agreed Finally, imitation consists of adjusting to the prices set on the price decline, they provided different explanations by other vendors. They gather information on sticker for this trend. Five of them argued that the reason for the prices by directly asking neighboring vendors. For downward trend was that the price of the box contain- instance, a vendor said: ing 520 stickers had reduced by approximately 30%. Two [Excerpt 2] “We set prices based on past sales expe- vendors claimed that prices lowered as a consequence rience and checking with other sellers.” of greater competition given the large number of ven- dors, whereas two other vendors claimed that demand One of the interviewed sellers claimed that sometimes had been decreasing. The remaining vendor claimed that they have to lower prices even if it leads to losses because price fluctuation was mostly associated with the perfor - their neighboring vendors create much competition. mance of the Colombian squad in the World Cup. More precisely, she says: We introduced first these two questions in which vendors would be more likely to acknowledge price variation than [Excerpt 3] “Well, I have tried to change the prices, when being asked about price changes that may hint discrim- because at least with the stickers, the shiny ones are ination based on buyers’ attributes. Then, we proceed to ask if the difficult ones and then I am the one that sets they charged different prices to males and females, and locals the highest prices. The rest of the stickers are cheap with respect to foreigners. All the respondents said that they because of the price war. It is about the one who sells charged the same prices regardless of gender. Similarly, all but more… is not the one who earns the most, but the one of the responders said that they charged the same prices one who sells the most… is the one who wants to fight regardless of nationality. The remaining vendor said: for a client, the one who gives him/her the lowest price. Also, clients put a price on the merchandise. [Excerpt 4] “Not really… the price is like the same. In other words, you cannot set the price because they What changes is as people say… as the saying goes… already come to impose the prices.” to the highest bidder, right?… then let’s say, those who are charged a little more, or something like that, The two vendors following an imitation strategy were maybe are the foreigners.” located on the cluster in the 147th Street, where vendors’ stalls are closer to each other than any other cluster, but The audit experiment and the interviews provide dif - ferent responses to whether sticker vendors discriminate against foreigners. The main challenge for interpreting the The official price per box to the public was 218,400 COP, about 21% more interviews’ responses is to disentangle between a social expensive. 6 Page 14 of 24 P. Zamora et al. desirability bias (i.e., vendors might believe that acknowl- geographic clusters, from 25,000 to 40,000 COP. Vendors edging price discrimination is objectionable) from a pat- also claimed that distributors started to give them similar tern of implicit (or unconscious) discrimination (Bertrand large sheets with multiple shiny foil stickers when they et al. 2005). An argument against the social desirability bias bought boxes containing packets of stickers. According to is that sellers acknowledge the trade of counterfeit stickers. a vendor, this was a compensation offered by the distribu - This argument is explored in-depth in the next section. tors, given the rarity of shiny foil stickers. Seven of the interviewed vendors claimed that they 5.3 Counterfeit stickers were aware of counterfeits, but they were not selling The rarity of shiny foil stickers in the 2018 World Cup them. The main reason exposed to not engage in the sale album gave origin to the presence of counterfeit cop- of counterfeits were reputational concerns. In most cases, ies among street vendors, as some of them confirmed. they do not want to lose their clientele since they have Regarding this rarity, one vendor said: been selling stickers in the same spot for at least three FIFA World Cups. Two of the vendors revealed that, [Excerpt 5] “In the last World Cup, there were sev- although they were not selling counterfeits, sometimes eral teams that came out a lot and others that did they exchanged stickers with buyers and, as a result of not. However, this year all the teams came out, as these exchanges, they ended up with counterfeits. The well as all the players. What did not come out were following is the testimonial from a vendor who acciden- the shiny ones. Let’s say, from a single box, there tally received counterfeits in an exchange: were 18, 19 shiny stickers coming out. There are 50 in total, considering the first and last page, as well [Excerpt 7] “Since I have the stall, I must sell what is as the emblems. Usually, 18 came out from a box, at original, but on the street, there are a lot of scanned most 20 or 22, you see?”. stickers that vendors are selling, and there are peo- ple who do not know. Then what happens? one day As a consequence of this rarity of shiny foil stickers, I had a problem with a DIAN official. He was a vendors confirm that people passed by offering them good client, and he brought colleagues here. It turns a set of 50 shiny foil counterfeit stickers. For instance, a out that I, by mistake, for being there (points out vendor said: to the street), and in the bookstore, I received some [Excerpt 6] “They pass by asking: < < Do you need counterfeit stickers, I exchanged them, and then I shiny stickers? > > , and I answer : < < Yes, but origi- was going to sell them. It turns out that precisely I nals > > , and then they say: < < Nooo, you know that gave them to him! and the official brought me the I don’t have those > > , so I respond: < < Ahm no! I police!… but I did not know, I did not know that the already have those, I need originals. > > They sell the stickers were fake. They are very well made, but nev - set of 50 shiny stickers for 25 (thousand COP) or 20 ertheless, there is a detail that you can recognize.” (thousand COP).” The remaining four vendors acknowledged that they were selling counterfeits. However, they sold both types Learning about the existence of counterfeit stickers was of stickers and claimed that they offered the originals one of our major motivations to conduct the interviews first. Only when buyers said the stickers were too expen - since price variation could reflect possession of counterfeits sive, they did offer counterfeits as a cheaper alterna - rather than price discrimination based on buyers’ attributes, tive. The following is the testimonial from one of these or even worse, a combination of both. Think, for instance, vendors: that the same price was quoted to locals and foreigners, but vendors offered original stickers to the former and counter - [Excerpt 8] “Look, I’m very honest with the client. I feits to the latter, then the two prices would not be compa- didn’t use to bring fake stickers, but clients asked me rable. We argue that this was not the case since counterfeits for fake ones… and now I bring fake ones. So I tell the appeared in the market, but they were only offered after client: < < Well, do you want cheap or do you want buyers claimed that shiny foil stickers were too expensive. expensive? Do you want the original or not? > > … Three different vendors mentioned that they started to be offered a large sheet containing the fifty shiny foil stickers in the collection two weeks before the kick-off A street vendor located in downtown said that the price of these 50 fake of the World Cup. The price of this sheet varied between stickers also fell over time like the original stickers:”they sell the package at 25,000, or 20,000. In the beginning, they sold it at 50,000.” With terms of trade highly advantageous for vendors, such as 3 for 1 sticker. However, buyers regularly accepted because they were close to com- plete the collection and had a large stock of duplicates. DIAN is the Colombian Tax and Customs National Authority. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 15 of 24 6 then, it depends on the customer. I tell them: And another one mentioned how investments were < < well, you are consciously buying a fake one > > . ” driven by expectations about the performance of the National squad four years before: The presence of counterfeits is not very likely to affect the outcomes in our audit experiment for two reasons. [Excerpt 10] “Four years ago, the jerseys were scarce. First, according to a couple of vendors, counterfeits were Colombia won, and the price went up… it immedi- easily detectable based on the lower brightness and the ately went up. So I said, this year, the same thing will lack of holograms, and it is therefore credible that they happen, and it was not true. I received a loan of 10 were not offered to buyers in our audit study. Second, our million pesos, and I invested it in a lot of jerseys. I script did not allow buyers to bargain over prices. There - paid 23,000 pesos (each), and today I went to buy fore, the typical offer of counterfeits as a response to a some small jackets for kids, and I realized that the price reduction was not likely to occur. (jerseys’) price now is 4,000 pesos… Too bad!… Now I can cry over spilled milk. If they win today, I go and buy. From there, I recover some of what I… It’s ter- 5.4 O ther insights from the stickers market rible, terrible. Do you know what that is? Look, we This last subsection is devoted to the analysis of three bought the jerseys for 23,000 pesos, and we are sell- additional lessons on how the stickers market works. ing them for 25,000. Can you imagine what it is like Some of these insights yield more general lessons regard- waiting for four years to make a profit of 2,000 from ing seasonal street-vending, primarily linked to sporting a jersey (0.7 US Dollars)?”. events. This testimonial also confirms the large investments in From the audit experiment and the interviews, we sporting events merchandise made by street vendors. Our learned that the stickers market is subject to multiple second insight is that these loans, combined with the long sources of uncertainty. The spatial distribution of ven - periods that street vendors have stayed in this business, dors, together with the low relocation costs in some suggest that (in this seasonal market) informality is a choice zones, revealed strategic uncertainty in the competition rather than an outcome associated with exclusion of the for clients. Additionally, the vendors also disclosed they formal labor force (Levy 2010; La Porta and Shleifer 2014). had made large investments and were concerned by the The ten interviewees that responded for how long they had rapid drop in prices of the merchandise associated with been street vendors during the World Cup season men- the World Cup. In fact, they mentioned that they were tioned on average 14.9  years. Nonetheless, some of them forced to decrease prices sooner than usual due to the have other off-season jobs different from street-vending. large competition that year and the number of counter- They take advantage of the flexibility from informal labor feit stickers circulating in the market. markets to work on the streets when it is more profitable. Another source of uncertainty we could identify from A final insight is how the costs of entry and exit from sea - the interviews was the connection between vendors’ sonal street-vending relate to the spatial location of vendors. profits and expectations and the performance of the In principle, if entry is not costly, vendors’ location is subject Colombian national squad in the World Cup. This link is to the “tragedy of the urban commons” (Webster 2007; Kim important since vendors invested in merchandise based 2012). Two reasons alleviate this problem. On the one hand, on the outcome of the Colombian national squad in the attractive selling spots are less scarce given the sparse loca- previous World Cup, which was exceptional. In addition, tion pattern nearby stores of a high-end supermarket chain. the performance during the 2018 FIFA World Cup not Besides, most street vendors do not see each other as fierce only shaped expectations about sales but also determined competitors. This is the testimonial of a vendor when asked the price of the merchandise, particularly of stickers. about her location nearby one of these stores: Vendors were afraid of an early elimination of the Colom- bian squad, which would drastically decrease the demand [Excerpt 11] “My spot was next to the door of Banco for goods related to the World Cup. The converse appears de Bogotá, and then I got sick and went home. There, to be also true for some vendors. For instance, a vendor on the sidewalk, there was a thing on the (parking) replied the following when asked about adjusting the blue zones, and I had a stall there, but then since I price of stickers: got cancer, I went home. Now I come here to work only during the World Cups [Interviewer: Ok… [Excerpt 9] “It depends… At least now during the and is it better to be around other vendors like this? World Cup, as long as Colombia wins, (prices) go Because generally there is a lot of people here] A lot! up.” Look, let me tell you one thing: what God has for you, he keeps it, no matter where you are.” 6 Page 16 of 24 P. Zamora et al. On the other hand, there are some areas where the rela- We combined the audit studies with eleven interviews tionship with policemen mediates location outcomes. In conducted with sticker vendors from the audited sample. most cases, entry costs might be higher since policemen ask The interviews shed light on crucial aspects to interpret the for merchandise (i.e., stickers or jerseys) to let them work on results from the audit studies. Interviewed vendors claim those areas. However, policemen might also give priority to that they do not discriminate based on buyer’s attributes. vendors established in a given spot for a  long time, under We argue that these responses might reflect implicit dis - some conditions. A vendor revealed that she has to work, crimination (Bertrand et al. 2005) rather than a social desir- once a week, on a spot where sales are good only in Decem- ability bias. Vendors concerned about providing socially ber (otherwise policemen might reassign her spot). She also desirable answers would have not naturally claimed that revealed that policemen transfer the costs of reducing occu- they sold counterfeits in addition to official stickers, as pation of the public space to vendors. This is her response: occurred with about half of them. From a more general perspective, the combination of audit studies with inter- [Excerpt 12] “Yes, (they let us work) under the condi- views allowed us to better understand the functioning of tion that we do not let any other new (vendor) get in” . this seasonal market. Although street vendors make large investments in merchandise related to the sporting event (including stickers), price variation is much better explained by the vendors’ location around the city and not by altering 6 Final discussion prices to specific buyers in a considerable amount. Street-vending of stickers is an informal market without The recent use of audit experiments in developing posted prices, which allows sellers to exercise a discre- countries introduces detectability as an important ele- tionary power. We conducted two audit studies on sticker ment for understanding behavior. For instance, Gross- vendors to test the existence of price discrimination in man and Honig (2017) show that rice sellers are more this context. Study 1 aimed at detecting price discrimi- likely to discriminate against High-class non-coethnics, nation based on two buyers’ attributes: nationality and not by charging directly a higher price, but by deliver- gender. We find evidence of price discrimination against ing smaller quantities, conditional on paying the same Argentinians with respect to the local Colombian experi- amount for a scoop of rice. Similarly, we found that price menters. Foreigners were charged on average 4.8% more discrimination based on nationality was most evident for per sticker, and the effects are significant for the most the expensive stickers, those in which price fluctuations expensive and the least expensive stickers. We argue that and scarcity can be used to argue discretionary prices. sellers use the higher price fluctuation of the most expen - This evidence calls for further studies focused on the sive ones, for which they can claim scarcity, as an excuse detectability of discrimination in informal markets. to increase prices for foreigners. For the least expensive Open questions include whether the subtleness of stickers, sellers infer that foreigners are less familiar with discrimination operates at a conscious or unconscious the low prices of regular stickers (i.e., those not printed in level, and whether the lower detection of discrimina- shiny foil), and is more likely that they receive quotations tion emerges in developing countries as a response to that are two or three times higher than the modal value. the fear of retaliation from the buyer. When formal By contrast, we did not find evidence of gender dis - and informal markets coexist, this question becomes crimination. Our script was designed to make sure that even more relevant since sanctioning mechanisms dif- any price difference between male and female buyers was fer between markets. If such fear of retaliation exists, not driven by a problem of second-order moral hazard it would be interesting to understand whether the per- (Balafoutas et al. 2015), in which vendors infer that men ceived threat is associated to direct punishment (i.e., quoted stickers for themselves, whereas women quoted verbal or physical aggression, or reports to the police) stickers for someone else. or indirect punishment (i.e., abstain from trade) (Bala- Study 2 aimed at detecting price discrimination based on foutas et al. 2014). Finally, fear of retaliation can lead to a good’s attribute: the length of the quoted sticker list. We models of multidimensional statistical discrimination, hypothesized that vendors could have inferred that buyers in which buyers’ (resp. vendors) attributes not only sig- having fewer stickers left to complete the collection would nal a willingness to pay (resp. to accept), but also indi- have a higher willingness to pay, which would have encour- cate the likelihood and expected damage of retaliation. aged them to raise the price of the same sticker when it belonged to a shorter list. We do not find evidence of this more sophisticated discriminatory behavior. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 17 of 24 6 Appendices Appendix 1 See Tables 6, 7, 8, 9 and Figures 2, 3, 4 We present three figures in this appendix. The first Appendix 2 two figures illustrate vendors’ location and their interac - Script for experimenters acting as vendors tion with experimenters acting as buyers. The left side of In the following scripts, squared brackets correspond to Fig.  2 depicts a map of the Northeastern side of Bogotá, instructions for the buyer that must not be read aloud. including the five clusters of sticker vendors. The right Sticker numbers separated by a slash (/) correspond to side of Figure 2 zooms in each one of the clusters to pro- the randomization of similar stickers between lists. The vide better information on vendors’ spatial location, text wrapped in < < > > is mentioned only by Argentinian Figure 3 shows a grid of buyers in the columns and ven- buyers. dors in the rows. The dots and diamonds indicate each vendor–buyer interaction registered in our experiment. Script #1: Long list [Before starting: Take out both the Figure  4 shows the cumulative distribution of quoted handwritten list and the printed list from the envelope. prices for the three least expensive stickers (i.e., those Check that the vendor’s ID matches the letters in the printed on regular paper), separately for Colombians and envelope. Mark all the information that is on the back of Argentinians. Table 6 Day on which each experimenter quoted prices and the route that followed Day Tueday June 5 Wednesday June 6 Thursday June 7 Friday June 8 Experimenter / Route COL-F-LL / North to COL-M-LL / South to North COL-F-SL / North to South ARG-F- South LL / North to South Experimenter Route ARG-M-LL / North to South COL-M-SL / South to North COL-F-LL refers to the Colombian woman in the Long List treatment. The COL-M-LL experimenter went back to Downtown during the morning on Friday, June 8, since there were scarce vendors during the morning on Wednesday, June 6. There was a large police presence in the area, which caused the vendors to disperse and made it difficult to find them. Also, the ARG-M-LL experimenter went back to Downtown during the morning on Thursday, June 7, since there were scarce vendors during the late afternoon the day before Table 7 Proportion of vendor–buyer interactions with unavailable stickers a Vendor did not have the sticker and did not give any price Vendors Stickers COL-F-LL ARG-M-LL COL-M-SL COL-F-LL ARG-M-LL COL-M-SL Downtown 0/21 1/12 1/6 0/189 6/108 1/24 Street 72nd 0/13 0/10 0/13 0/117 0/90 0/52 Street 85th 0/9 1/9 1/7 0/81 8/81 1/28 Street 93rd 1/7 0/8 1/7 6/63 0/72 1/28 Street 147th 0/8 0/7 0/7 0/72 0/63 0/28 Total 1/58 = 1.72% 2/46 = 4.35% 3/40 = 7.5% 6/522 = 1.15% 14/414 = 3.38% 3/160 = 1.87% b Vendor did not have the sticker but did mention a price Vendors Stickers COL-F-LL ARG-M-LL COL-M-SL COL-F-LL ARG-M-LL COL-M-SL Downtown 0/21 1/12 1/6 0/189 4/108 1/24 Street 72nd 0/13 1/10 2/13 0/117 3/90 3/52 Street 85th 0/9 0/9 0/7 0/81 0/81 0/28 Street 93rd 1/7 1/8 1/7 3/63 6/72 1/28 Street 147th 1/8 0/7 0/7 1/72 0/63 0/28 Total 2/58 = 3.45% 3/46 = 6.52% 4/40 = 10% 4/522 = 0.77% 13/414 = 3.14% 5/160 = 3.12% Columns correspond to the three buyers who collected this information in detail. COL-F-LL refers to the Colombian woman in the Long List treatment. For the Long List, the number of missing values in bold font coincide with the number of missing values in the database for the experimenter. For the Short List, even though nine vendors did have the four stickers, they only mentioned the price for the full list, and not for each of them individually 6 Page 18 of 24 P. Zamora et al. Table 8. Balance across attributes made salient in Studies 1 and 2. Coefficients correspond to linear probability models. (1) (2) (3) (4) Sample Long List Short List Full Sample Dep. Variable Gender (=1 if Male Buyer) Nationality (=1 if Foreigner) Gender (=1 if Male Buyer) List Length (=1 if Long List) (Perceived) Vendor’s age (= 0.232*** (0.072) − 0.040 (0.076) − 0.056 (0.105) 0.024 (0.076) 1 if 35) Vendor’s gender (=1 if 0.032 (0.073) − 0.046 (0.075) − 0.068 (0.106) − 0.018 (0.076) Male) Isolated vendor 0.094 (0.075) 0.008 (0.077) 0.026 (0.115) − 0.084 (0.076) Journey (=1 if Afternoon) − 0.129 (0.092) − 0.010 (0.091) 0.379*** (0.112) 0.170** (0.085) Constant 0.373*** (0.099) 0.522*** (0.101) 0.262*** (0.117) 0.454*** (0.092) Observations 199 199 88 194 R-squared 0.058 0.003 0.129 0.024 *p < .1, **p < .05, ***p < .01. Robust standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in thousands of Colombian pesos (kCOP). The sample size in model (4) is 194 observations because Foreign buyers were excluded from the regression ask her for the price that she would charge for it if she Table 9 Eec ff t of vendor–buyer gender interactions on the price finds it for you. If the vendor says she could not find it, quoted by the vendor ask for the price she charged the last time she sold it.] Okay, let’s start with the team emblems. I need three. Variables (1) Price [kCOP] (2) Price [kCOP] First of all, I need the emblem of my country: Colom Foreigner 0.203* (0.113) 0.203* (0.121) bia/Argentina. It’s number 620/260. How much is it? Male Buyer x Female Vendor 0.336 (0.312) 0.336 (0.381) I also need the emblem of Argentina/Colombia. It’s Female Buyer x Male Vendor 0.153 (0.166) 0.153 (0.178) number 260/620. How much is it? And the emblem of Male Buyer x Male Vendor − 0.097 (0.155) − 0.097 (0.157) Peru/Uruguay. It’s number 220/80. How much is it? Constant − 0.200 (0.330) − 0.200 (0.288) I am missing one legend, < < the best player in his R-squared 0.799 0.799 tor y, > > Maradona . It’s number 664. Controls and Vendor FE Yes Yes < < The guy is a genius! > > How much is it? Errors Standard Robust Okay. Let’s continue from the beginning. I need two Number of observations 1712 1712 from the front page. *p < .1. Standard errors are shown in parentheses. The dependent variable in all Number 2, which is the Trophy; and number 7, which specifications is the vendor’s price quoted in thousands of Colombian pesos is the Ball. How much for each one? I also need one sta (kCOP). Controls in all regressions: geographic cluster, day and sticker fixed effects, (perceived) age and gender of the vendor, whether the vendor has dium, number 11/12. How much is it? nearby vendors, and journey I need (the picture of ) one team, it is Denmark/Ice- land, number 241/281. How much is it? And finally, I need one player, number 129/574. It’s a the printed list. Hide the printed list and do not get too player from Spain/England. How much is it? close to the vendor.] [Annotate all the prices in the hand-written list and Good morning/afternoon, how are you? terminate the interaction.] My husband/wife and I are looking for the missing Thank you! I may come back, or my husband/wife will stickers to complete our kid’s PANINI album. come back. At what time of the day can I find you here? He/She is also looking for the stickers right now, so Thanks again, have a good day. for the moment, I am only quoting. Do you complete [Walk away from the seller and write all the prices on lists? the printed list.] [If the vendor says “Yes,” continue the script. Other- wise, say goodbye and look for the next vendor ID.] Script #2: Short list Since the protocol is very similar, we OK, I tell you which ones I need, and you tell me the only provide below the part of the script concerning the price. quotation of the four stickers in the Short List treatment. I am going to write them down because they are for Unlike the Long List treatment, in this case we quote stick my kid, and I also have to talk to my husband/wife. ers in ascending order according to the album’s numbers. [Start writing down the prices on the list assigned We altered the order because it would be easier for the to that vendor. If the vendor does not have a sticker, Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 19 of 24 6 Fig. 2 Sampling of sticker vendors in five clusters in Bogotá. The maximum distance between clusters is 15 km. Panels on the right side represent an 800 × 800 m grid for each cluster 6 Page 20 of 24 P. Zamora et al. 59 Vendors, 287 Obs. C-24 C-23 C-22 C-21 C-19 C-18 C-16 C-15 C-14 C-13 C-12 C-11 C-10 C-9 C-8 C-7 C-6 C-5 C-4 C-3 C-1 S-14 S-13 S-12 S-11 S-10 S-9 S-8 S-7 S-6 S-4 S-3 S-2 S-1 T-9 T-8 T-7 T-6 T-5 T-4 T-3 T-2 T-1 P-11 P-9 P-8 P-7 P-6 P-5 P-3 P-1 K-13 K-11 K-10 K-7 K-3 K-2 K-1 N-1 1. Arg-F-LL 2. Arg-M-LL 3.Col-F-LL 4. Col-M-LL 5. Col-F-SL 6. Col-M-SL LL (199 obs.) SL (88 obs.) Fig. 3 Mapping between buyers and vendors. Columns represent each one of the buyers, and rows represent each one of the sellers. Circles denote a vendor–buyer interaction in the Long List treatment, and hollow diamonds denote a vendor–buyer interaction in the Short List treatment. vendor to look for the stickers, and in the Short List treat- I need the Panini Knight Logo/FIFA Fair Play, it’s num- ment, there was no need to prime the buyer’s nationality. ber 0/1. How much is it? I also need the Trophy, number Okay, let’s start with the emblems on the first page. 2. How much is it? Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 21 of 24 6 Stadium Team Player 100 100 100 80 80 80 60 60 60 40 40 40 20 20 20 0 0 0 0 1 2 3 0 1 2 3 0 1 2 3 kCOP kCOP kCOP Colombians Argentinians Fig. 4 Cumulative distribution of quoted prices. Panels from left to right represent the cumulative distributions for the stickers with stadiums, teams, and players. The distribution for Colombians is displayed in blue dots. The distribution for Argentinians is displayed in red triangles. Good. I also need the logo of Brazil/Germany, it’s num- • Do you share your business with someone else? Fam- ber 340/420. How much is it? Finally, I need the logo of ily? Partner? Friend? Employee? Colombia. How much is it? • At what time do you sell more? • How did you establish your prices? • Do you get upset if people tell you that they are only Appendix 3 going to quote? See Table 10. • Do you change the prices throughout the day? • Do you charge more per sticker to some people than Questions for the structured interview others? Women with respect to men? Foreigners with Questions in parenthesis were only asked when applied. respect to locals? To young people or to older peo- ple? • What is your usual work schedule? Why that sched- • Any difference in prices before it started and during ule? the World Cup? • Does anyone else work here with you? Are you usu- ally accompanied? Table 10 Schedule of conducted interviews Date Number of Interactions Time Zone Gender Perceived age 6 1:30 p.m Street 93rd Female Old woman July 2, 2018 (holiday) 6 2:00 p.m Street 93rd Male Young man 6 3:00 p.m Street 147th Female Middle Age 6 3:30 p.m Street 147th Male Middle Age 6 9:50 a.m Downtown Male Middle Age 5 10:15 a.m Downtown Female Old woman July 3, 2018 (business day) 6 11:10 a.m Street 72nd Female Middle Age 6 11:40 a.m Street 72nd Male Old man 6 12:00 p.m Street 85th Male Middle Age 6 12:10 p.m Street 85th Male Middle Age July 13, 2018 (business 4:30 p.m Street 147th Female Middle Age day) Percent 6 Page 22 of 24 P. Zamora et al. • When were the highest prices reached and when el dicho... al mejor postor, ¿no?... entonces pues diga- were the lowest? mos, a los que se les cobra un poquito más o algo así, • Do you sell or used to sell counterfeit stickers? de pronto sí son los extranjeros.” • How can you find the counterfeits? Who sells them? [Excerpt 5] “En el mundial pasado había varios • Why are you located here? equipos que salían mucho y otros que no salían. En • (Why are you so close to the supermarket?) cambio este año salieron todos los equipos, todos los • (If you know there are many vendors here, why not jugadores. Lo que no salieron fue las doradas. Diga- go further?) mos, de una sola caja, estaban saliendo por ahí 18, • For how long have you established your business 19 escudos. Son 50 en total, entre primera, última here? página y escudos. Salían 18, por mucho salían 20, • What do you do outside the World Cup season? Do 22, ¿si ves?” you sell something else? [Excerpt 6] “Pasan preguntando: “necesitan escu- • Does the police often come to carry out inspections? dos?”, yo contesto: “sí, pero originales”, entonces ellos What do you do when the police arrive? dicen: “Nooo, usted sabe que yo no tengo de esos”, • Do you pay, or paid before, social security? Pension entonces respondo: “¡ahm no!, yo ya tengo de esos, contributions? Health insurance? necesito originales”. Eso venden el paquete de 50 • Have you ever rented a place to sell your merchan- láminas a 25 o a 20” dise? (How much do you pay for rent?) [Excerpt 7] “A mí por el local me toca vender sólo lo • Why did you decide to dedicate yourself to selling que es original, pero en la calle hay mucha lámina things in this way? escaneada que los vendedores están ofreciendo y hay • Do you live far away? gente que no conoce. Entonces, ¿qué pasa? a mí un • What do you do with the merchandise after the jour- día me pasó un inconveniente con un funcionario ney? Where do you leave it? de la DIAN. Era un buen cliente y é me traía com- pañeros. Resulta que yo por equivocación y por estar allá (señala la calle) y estar en la librería, recibí Original Excerpts in Spanish unas láminas y yo las intercambié y después las iba a vender. ¡Resulta que precisamente se las di a é! y el [Excerpt 1]“Pues es que ya uno tiene un recorrido funcionario ¡me trajo la policía!... pero yo no sabia, con el mundial, ¿sí?... que por lo menos uno coge y yo no sabía que las láminas eran falsas. Es que son destapa cajas… porque es que uno destapa... tiene muy bien hechas, pero sin embargo tiene un detalle es que destapar cajas o pacas como se dice... porque que uno puede reconocer.” una paca trae 12 cajas. Entonces pues digamos uno [Excerpt 8] “Mire yo soy muy honesta con el cliente. destapa 3, 4 pacas. Entonces, ya arreglando las Yo acá no trabaja las chiviadas, pero a mí me pedían fichas por números y todo, ahí uno deduce cuál es chiviadas... y ahora yo traigo chiviadas. Entonces yo el equipo que más sale, los jugadores que más salen, le digo al cliente: < < bueno, quiere barato o quiere las fichas que menos salen. Entonces de ahí los que caro?, ¿quiere original o no original? > > ... entonces más saca son los más baratos y los que menos salen ya depende del cliente. Le digo: < < bueno, ya a con- son los má costosos.” ciencia lleva chiviada... > > ” [Excerpt 2] “Establecimos los precios de acuerdo a la [Excerpt 9] “Es dependiendo... por lo menos ahorita experiencia de ventas anteriores y averiguando con en el mundial, a medida que va ganando Colombia, otros vendedores.” va subiendo.” [Excerpt 3] “Pues yo he tratado de cambiar los pre- [Excerpt 10] “Hace cuatro años las camisetas estu- cios, porque por lo menos con las láminas los escu- vieron escasas. Ganó Colombia y se subió el precio... dos son los dif íciles y pues ahí yo soy la que má caro enseguida subió el precio. Entonces yo dije, este año vende. El resto sí por la guerra de precios, entonces va a pasar lo mismo, y mentiras. A mí me hicieron se venden barato. Es el que más vende… no es el un préstamo de 10 millones de pesos y lo invertí que más gane, sino el que más vende... el que quiera en un poco de camisetas. La pagué a 23,000 pesos pelear a un cliente, el que le dé más barato, aparte y hoy fui a comprar unas chaqueticas para niño y que los clientes le ponen precio a la mercancía... o me doy cuenta que está a 4,000 pesos... ¡ya qué!... ya sea uno ya no puede ponerle precio, porque ellos puedo llorar sobre la leche derramada. Si ganan voy vienen ya a ponerle el precio a uno.” y compro. Ahí repongo algo de lo que... terrible... ter- [Excerpt 4] “No, de por sí, no... el precio es como el rible... ¿sabe qué es eso? mire, compramos camiseta mismo. Lo que cambia es como se dice... como dice a 23,000 pesos y la estamos vendiendo a 25,000... ¿se Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 23 of 24 6 Ayres, I., Siegelman, P.: Race and gender discrimination in bargaining for imagina sumercé esperar cuatro años para ganarse anew car. Am Econ Rev 85, 304–321 (1995) uno en una camiseta 2,000 pesos?” Azmat, G., Petrongolo, B.: Gender and the labor market: What have we [Excerpt 11] “Mi puesto era en la puerta del Banco learned from field and lab experiments? Labour Econ 30, 32–40 (2014) Baert, S., De Pauw, A.-S.: Is ethnic discrimination due to distaste or statistics? Bogotá y me enferme´ y me fui para la casa. Ahí en Econ Lett 125(2), 270–273 (2014) la bahía había una cosa de las zonas azules y ahí Balafoutas, L., Kerschbamer, R., Sutter, M.: Second-degree moral hazard in a yo tenía un puesto, después como me dio cáncer real-world credence goods market. Econ J. 127(599), 1–18 (2015) Balafoutas, L., Nikiforakis, N., Rockenbach, B.: Direct and indirect punishment me fui pa’ la casa. Ahora no vengo sino a traba- among strangers in the field. Proc Natl Acad Sci. 111(45), 15924–15927 jar los mundiales acá. [Entrevistadora: ¿y es mejor (2014) estar así rodeado? Porque aquí generalmente hay Baldini, M., Federici, M.: Ethnic discrimination in the Italian rental housing market. J Hous Econ. 20(1), 1–14 (2011) más gente, ¿no?] ¡Muchísima! Mire sumercé, yo Balsa, A.I., McGuire, T.G.: Statistical discrimination in health care. J Health le digo una cosa: lo que Dios le tiene a uno, se lo Econ 20(6), 881–907 (2001) guarda, así esté uno donde esté.” Baranzini, A., Schaerer, C., Ramirez, J.V., Thalmann, P.: Do foreigners pay higher rents for the same quality of housing in Geneva and Zurich? [Excerpt 12] “Sí, nos dejan trabajar aquí con la Swiss J Econ Statist 144(4), 703–730 (2008) condición de que no dejemos meter a ninguno Becker, G.S.: The economics of discrimination. University of Chicago Press, nuevo.‘” Chicago (1957) Bertrand, M., Chugh, D., Mullainathan, S.: Implicit discrimination. Am Econ Rev 95(2), 94–98 (2005) Acknowledgements Bertrand, M., Mullainathan, S.: Are Emily and Greg more employable than We gratefully acknowledge support from Santiago Sautua, Yuliet Verbel, and Lakisha and Jamal? A field experiment on labor market discrimination. Juan Baquero during the execution of the field experiment. We also thank Am Econ Rev 94(4), 991–1013 (2004) Diego Aycinena, Francesco Bogliacino, Simeon Schachtele, and participants in Castillo, M., Petrie, R., Torero, M., Vesterlund, L.: Gender differences in bar - the LACEA-BRAIN workshop for their valuable comments. gaining outcomes: a field experiment on discrimination. J Public Econ The information and opinions presented herein are entirely those of the 99, 35–48 (2013) authors. Financial Support from the program “Inclusión productiva y social: Charness, Gary, and Peter Kuhn. 2011. "Lab Labor: What Can Labor Econo- programas y políticas para la promoción de una economía formal, código mists Learn from the Lab?" In Handbook of Labor Economics , Vol. 4, 60185, que conforma la Alianza EFI, bajo el Contrato de Recuperación Contin- ed. Orley Ashenfelter and David Card, 229-330. Amsterdam: Elsevier. gente No. FP44842-220-2018.” is gratefully acknowledged North-Holland. Chen, L.-Z., Wei-Min, Hu., Szulga, R., Zhou, X.: Demographics, gender and Authors’ contributions local knowledge—price discrimination in China’s car market. Econ Lett All authors have participated in (a) conception, design, and analysis of the 163, 172–174 (2018) data; (b) drafting the article or revising it critically for important intellectual Csizma, K.A., Wittig, A.F., Schurr, K.T.: Sport stereotypes and gender. J Sport content; and (c) approval of the final version. All authors read and approved Exerc Psychol 10(1), 62–74 (1988) the final manuscript. Currie, J., Lin, W., Zhang, W.: Patient knowledge and antibiotic abuse: evidence from an audit study in China. J Health Econ 30(5), 933–949 Funding (2011) Financial Support from the program “Inclusión productiva y social: programas y Fang, Hanming, and Andrea Moro. 2011. “Theories of Statistical Discrimination políticas para la promoción de una economía formal, código 60185, que conforma and Affirmative Action: A Survey.” Pp. 133–200 in Handbook of Social la Alianza EFI, bajo el Contrato de Recuperación Contingente No. FP44842-220- Economics, Vol. 1A, edited by J. Benhabib, A. Bisin, and M. O. Jackson. 2018.” is gratefully acknowledged. Amsterdam: North-Holland, Elsevier. Flage, A.: Ethnic and gender discrimination in the rental housing market: Evi- Availability of data and materials dence from a meta-analysis of correspondence tests, 2006–2017. J. Hous. Data and code are available in an Open Science Foundation repository: https Econ. 41, 251–273 (2018) ://osf.io/gzhfs /. GrabowKu¨ hl, H.M.: You don’t bend it like beckham if you’re female and reminded of it: stereotype threat among female football players. Front Competing interests Psychol 10, 1963 (2019) The authors have no affiliation with any organization with a direct or indirect Grossman, S., Honig, D.: Evidence from Lagos on discrimination across ethnic financial interest in the subject matter discussed in the manuscript. and class identities in informal trade. World Dev. 96, 520–528 (2017) Grytten, J., Skau, I., Sørensen, R.: Do expert patients get better treatment than Author details others? Agency discrimination and statistical discrimination in obstetrics. Department of Economics, Texas A&M University, College Station, TX, USA. J Health Econ 30(1), 163–180 (2011) Department of Economics, Universidad del Rosario, Bogotá, Colombia. Kim, A.M.: The mixed-use sidewalk: Vending and property rights in public space. J Am Plan Assoc 78(3), 225–238 (2012) Received: 30 October 2019 Accepted: 12 January 2021 Kim, S.-H., Fernanda, L., de Leon, L.: In-group and out-group biases in the marketplace: a field experiment during the World Cup. Oxford Economic Papers 71(3), 528–547 (2018) La Rafael , P., Shleifer , A.: Informality and development. J Econ Perspect 28(3), 109–126 (2014) References Levy, S.: Good intentions, bad outcomes: social policy, informality, and eco- Apollo, M.: Dual pricing-two points of view (Citizen and Non-citizen) case of nomic growth in Mexico. Brookings Institution Press, Washington (2010) entrance fees in tourist facilities in Nepal. Procedia Soc Behav Sci 120, List, J.A.: The nature and extent of discrimination in the marketplace: Evidence 414–422 (2014) from the field. Quart J Econ. 119(1), 49–89 (2004) Arrow, Kenneth. 1973. The theory of discrimination in labor markets. In: O. List, J.A., Rasul, I.: Field experiments in labor economics. In: Handbook of labor Ashenfelter, A. Rees (Eds.). Princeton, NJ: Princeton University Press. economics, vol. 4, pp. 103–228. Elsevier, Amsterdam (2011) Auspurg, K., Hinz, T., Schmid, L.: Contexts and conditions of ethnic discrimi- Martínez, L., Short, J.R., Estrada, D.: The urban informal economy: Street ven- nation: Evidence from a field experiment in a German housing market. dors in Cali, Colombia. Cities 66, 34–43 (2017) J Hous Econ. 35, 26–36 (2017) 6 Page 24 of 24 P. Zamora et al. Mullainathan, S., Noeth, M., Schoar, A.: The market for financial advice: an audit Webster, C.: Property rights, public space and urban design. Town Plan Rev 78, study. National Bureau of Economic Research, Cambridge (2012) 81–101 (2007) Oxford, S., McLachlan, F.: “You Have to Play Like a Man, But Still be a Woman”: Yip, C.M., Wong, R.-K.: Gender-oriented statistical discrimination theory: Empiri- young female colombians negotiating gender through participation in cal evidence from the Hong Kong labor market. Res Soc Stratif Mob 37, a sport for development and peace (SDP) organization. Sociol Sport J 43–59 (2014) 35(3), 258–267 (2018) Siegelman, P., Heckman, J.: The urban institute audit studies: their methods Publisher’s Note and findings. Clear Convinc Evid 187, 258 (1993) Springer Nature remains neutral with regard to jurisdictional claims in pub- Turchin, Peter. 2016. Ultrasociety: How 10,000 Years of War Made Humans the lished maps and institutional affiliations. Greatest Cooperators on Earth. Chaplin, CT: Beresta Books. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal for Labour Market Research Springer Journals

Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018 FIFA World Cup

Loading next page...
 
/lp/springer-journals/price-discrimination-in-informal-labor-markets-in-bogot-an-audit-Zsx0v0DiQn
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
ISSN
1614-3485
eISSN
2510-5027
DOI
10.1186/s12651-021-00285-1
Publisher site
See Article on Publisher Site

Abstract

We conducted an audit experiment to examine whether street vendors in Bogotá (Colombia) exert price discrimina- tion based on buyers’ attributes, such as gender and nationality, and based on product characteristics, such as the increasing marginal valuation of items needed to complete a collection. We exploited the seasonal demand for album stickers related to the FIFA World Cup Russia 2018. In our within-subjects design, experimenters carried out in-person audits and quoted a pre-determined list of missing stickers. They interacted with 59 sticker vendors located in five geographic clusters and collected 287 vendor–buyer interactions. We find that prices quoted to foreign buyers are higher than prices quoted to Colombian buyers. By contrast, we do neither find evidence supporting direct gender- based discrimination, nor that vendors charge a higher price per sticker when the list of missing stickers is shorter. We complement the study with a qualitative analysis based on interviews that reveal vendors’ pricing strategies, their awareness of price discrimination, and the trade of counterfeits. The qualitative results suggest that price discrimina- tion appears to be unconscious. Keywords: Colombia, Dual labor markets, Football, Latin america, Sports, Street vendors JEL Classification: C93, J46 1 Introduction For a season related to a sports competition, such as the FIFA World Cup, which we explore in this study, vendors The coexistence of formal and informal labor markets is prefer to have an accelerated turnover of merchandise prevalent in low and middle-income countries. In par- instead of storing it, since it is a very costly option. There ticular, street vending accounts for a significant part of - their informal labor sector since it is a major source of fore, in the absence of posted prices in informal markets, employment and income in the main cities. Some street we explore whether price discrimination appears as a vendors sell the same products year-round, while oth- possible mechanism, either conscious or unconscious, ers invest in seasonal products associated to festivities employed by street vendors to quickly recover their (e.g., Christmas and Halloween) and sports competitions. investments. Selling this type of merchandise potentially comes with We explore price discrimination in the market of stick- greater returns, but it is also tightly connected to higher ers for the FIFA World Cup’s album. The tradition of economic risks. Once the season is over, vendors have completing this collection has persisted over decades to drastically reduce prices to clear stocks or store them in Colombia, despite the arrival of digital information for extended periods, which implies additional costs. to keep track of competitors (teams and players) in the World Cup. Among the different seasonal products asso - ciated with the World Cup, such as jerseys and flags, we *Correspondence: cesar.mantilla@urosario.edu.co opted to use stickers since they provide two advantages. Department of Economics, Universidad del Rosario, Bogotá, Colombia First, we can test different discriminatory strategies Full list of author information is available at the end of the article © The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. 6 Page 2 of 24 P. Zamora et al. since we are able to manipulate the stickers’ salience complete the collection, she might charge a larger price individually. Second, the buyer can credibly signal a low for the same sticker when it belongs to a shorter list, as likelihood of future interactions because buyers mostly if she anticipates a buyer’s higher valuation for the last recur to street vendors when opportunities to trade with stickers left. other collectors have been exhausted, and the purchase The literature on market discrimination has mainly of sealed packets of stickers is likely to bring only dupli- focused on two types of discrimination. One of them cates. That is, buyers mostly interact with street vendors is taste-based discrimination, related to a preference- to obtain the last missing stickers in the collection. This driven mechanism (Becker 1957). The other is statisti - last feature allows us to minimize reputational concerns cal discrimination, a belief-driven mechanism relying on related to the intention of suppliers to offer lower prices self-fulfilling stereotypes (Arrow 1973). Whereas trans- to ensure future sales. actions data provide evidence of discrimination in mar- We conduct two studies to explore three potential kets associated with labor, health, housing, tourism, and sources of discrimination. In the first study, we evalu - retail sales (Apollo 2014; Balsa and McGuire 2001; Baran- ate two sources of discrimination that might be inferred zini et  al. 2008; Chen et  al. 2018; Grytten et  al. 2011; directly from the buyer’s appearance and accent. These Yip and Wong 2014), the controlled environment from are gender and nationality. We have four experimenters experiments has been useful for pointing out more spe- quoting a list of nine stickers. Two of them are Colom- cific discrimination mechanisms (Ayres and Siegelman bian and two of them are Argentinian, and there are a 1995; Auspurg et al. 2017; Baert and De Pauw 2014; Bal- woman and a man corresponding to each nationality. dini and Federici 2011; Castillo et al. 2013; Flage 2018). In the second study, we evaluate the third source of dis- Conducting an experiment involves an implicit trade- crimination. It relates to the quoted set of stickers (i.e., off between control and external validity. On the one the length of the stickers’ list). We have four Colombian hand, an experiment closer to the laboratory setting experimenters. Two of them quote a long list of nine provides greater control, making it easier to test spe- stickers, and two of them quote a short list of only four. cific mechanisms related to discrimination. On the other There are one woman and one man corresponding to hand, natural experiments emphasize realism while hold- each type of list. ing sufficient control to detect discrimination patterns Discrimination based on nationality is worth being (Charness and Kuhn 2011; List and Rasul 2011). Corre- explored in this context for two reasons. First, since spondence studies, within the family of natural experi- sports competitions increase national identity salience ments, aim to compare labor outcomes between testers (Turchin, 2016), it allows to test discrimination based (“confederates” or “auditors”) that, by assumption, differ on in-group favoritism and out-group hostility. Dis- only in the characteristic of interest for the analysis of crimination based on group boundaries has been tested discrimination, which is typically gender (Bertrand and in the context of a previous FIFA World Cup in a mar- Mullainathan 2004; Azmat and Petrongolo 2014). ket for low-cost electronic devices in Brazil (Kim et  al. The audit study technique employed in this paper is 2018). Second, foreigners might be subject to discrimi- methodologically close to correspondence studies, and nation based on the sellers’ belief that willingness to pay it is based on controlling the characteristics of the test- is higher for foreigners than locals. For developing coun- ers on one side of the market. When interacting with the tries with an increasing tourism industry, it is essential other side of the market, it aims to match most of the to understand whether informal markets, particularly attributes between testers except the one of interest for street-vending, are discriminatory against foreigners and detecting discrimination. In correspondence studies, the thus have reputational consequences for the whole tour- analysis focuses on the job-offer stage and may include ism industry. prices (List 2004) and financial or medical advice (Mul - We are also interested in gender discrimination given lainathan et  al. 2012; Currie et  al. 2011). In our audit, the prevalent stereotypes associating masculinity and there is a broader set of outcomes. In the experiment, sports competitions, particularly in Colombia (Csizma testers act as potential buyers and quote a set of prede- et  al. 1988; Grabow and Kühl 2019; Oxford and McLa- fined stickers, while street vendors provide prices in their chlan 2018). Sellers might assume that women have natural environment (remaining unaware of the audit). less experience buying stickers or are less knowledge- able about stickers’ prices and, therefore, discriminate accordingly. The third element useful to analyze price discrimi - Recent models bring other mechanisms to the discussion, including unin- nation is the length of the quoted list. Since the vendor tentional (or implicit) discrimination, search frictions in the labor market, and knows that the buyer is quoting the few stickers left to endogenous group formation (Bertrand et al. 2005; Fang and Moro 2011). Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 3 of 24 6 We contribute to this literature by conducting an is, two to three months in advance of the season’s high audit experiment in the context of an informal market, peak. where the lack of enforcement on mandatory posted In 2018, Completing the collection of 670 stickers cost prices increases sellers’ discretionary power (Grossman at least 36% of the monthly minimum wage in Colom- and Honig 2017). We find evidence of price discrimina - bia. Hence, collectors tend to be located in the city’s eco- tion based on the buyer’s nationality: vendors charge on nomic centers and high socio-economic residential areas. average 4.8% more per sticker to Argentinians, and the Vendors react accordingly and are located in areas with a effects are significant for the most expensive and the least large influx of office workers, forming clusters of vendors, expensive ones. Our conjecture for this finding is that and in the surroundings of stores of a high-end super- expensive stickers have greater price fluctuation, mak - market chain. ing it easier for vendors to hide their intention to quote Depending on their budget and time constraints, col- higher prices. Regarding the least expensive stickers, lectors recur to different strategies when buying sealed some vendors appear to believe that foreigners are less packets becomes too expensive due to the frequency of familiar with the low price charged for stickers printed duplicates. One alternative is exchanging stickers at parks in regular paper (instead of shiny foil). By contrast, we or malls, where gatherings coordinated via social media do not find evidence of price discrimination based on occur. However, they can be time-consuming when the buyers’ gender, and neither on the length of the quoted stock of duplicates is not very large. Another alternative sticker list. for collectors is to interact directly with the vendors. The We complement this study with a qualitative analysis advantage is that they tend to have a much larger set of based on eleven interviews with street vendors. We focus stickers to sell (and sometimes trade, although in much on describing their pricing strategies, their awareness of more favorable exchange terms for vendors). Therefore, price discrimination, and their perception (and trade) collectors whose budget constraints are looser prefer this of counterfeit stickers. The results from the interviews strategy. reveal that vendors do not seem to discriminate con- The price charged on the streets for stickers printed in sciously. Ten out of eleven interviewees state that they regular paper (players, pictures of teams, and stadiums) do not charge higher prices to foreigners, whereas about dwells between 1.25 and 2 times the sticker’s unit cost half of them admit selling counterfeit stickers. This result in a packet. By contrast, special stickers printed in shiny weakens the alternative argument that a social desirabil- foil tend to have a much larger price. For the 2018 World ity bias drives their responses to price discrimination. Cup, the fifty shiny foil emblems were scarce. Conse - quently, the price charged by sticker vendors reached 2 Experimental setting between 15 and 50 times the sticker’s unit cost in a 2.1 Description of the stick ers market packet. An alternative strategy was to buy the last miss- Street vendors, who represent a significant part of the ing stickers directly from the manufacturer. However, informal labor force in Bogotá, Colombia, are not mar- the stickers that could be purchased under this strategy ginalized and typically can afford the acquisition of sea - became unavailable in Colombia one month before the sonal products (Martínez et al. 2017). During the Football World Cup kick-off (three weeks before our data collec - World Cup, these products include jerseys and flags, and tion process). As in many other street-vending activities, stickers sold individually. The sale of individual stickers sticker vendors are used to bargaining prices with poten- is popular among street vendors because they are aware tial customers and have fast interactions that do not lead of the high costs of completing the collection and the to a transaction in a considerable number of cases. associated willingness to pay for individual stickers from a considerable number of collectors. Since vendors have high expectations about the World Cup season, they aim to accrue several duplicates of each sticker beforehand to While the collection only includes 670 stickers, the stock of some vendors reached about 10,000 stickers. respond to future demand. Vendors usually start open- A collector would need to buy 134 packets containing five stickers each, ing boxes as soon as the producer releases stickers. That for a total price of 281,400 COP. The monthly minimum wage for 2018 was 781,242 COP. This amount is a lower bound for the cost, under the implau- sible assumption that the collector would trade any duplicate she obtains. At the time of the experiment, $1 US Dollar was equivalent to (approxi- mately) 2,900 COP. 2 5 To give an idea on the high valuation of stickers, the 2018 census on inven- A Colombian newspaper issued an article entitled “Panini will no longer tory reduction and theft prevention (Censo Nacional de Mermas y Prevención ship stickers for the Russia’s World Cup album” (https ://perma .cc/X34B- de Pérdidas) revealed that sticker packets were the fifth most stolen item from DHGE). Unlike in the European Union and the United States, collectors supermarket chains in the whole year (https ://perma .cc/KGM4-MHAH), cannot directly purchase the complete sticker collection from the official after canned food products, soft-drinks, candies, and toiletries. distributor’s website. 6 Page 4 of 24 P. Zamora et al. 2.2 F ield site and vendors’ selection We avoided buyers’ overlap in the same cluster by delay- We exploit the seasonal demand for album stickers of ing the data collection from one of the experimenters. the Football World Cup 2018. Six experimenters acted Table 6 (see Appendix 1) details the list of days in which as potential buyers and approached fifty-nine street each experimenter quoted the prices. vendors. We define as vendor any subject, or group of subjects, who sell stickers in a fixed or mobile stall. 2.3 Execution of the audits Regardless of the type of stall, vendors’ locations are Members from the research team collected the quoted fixed. Hence, two or more subjects who share a stall are prices in pairs. One of the team members who were ini- treated as a unique vendor. In consequence, it is possi- tially in charge of mapping vendors’ locations always ble that two different buyers obtained the prices from the accompanied the buyer. In this way, the companion team same vendor ID but interacted with different subjects. member could verify that the buyer was approaching The vendors included in our sample were located in the right vendor. Moreover, we only made contact with five different clusters along the Eastern side of the city, the vendor when there were no potential buyers around. between Street 13th (downtown) and Street 147th. These The buyer approached the vendor and asked for a list of clusters coincide with centers of economic activity or res- stickers’ prices, while the companion waited at a prudent idential areas of medium–high and high socio-economic distance. The opening line of the script, leading to the status. Other areas of the city were not sampled because vendor–buyer interaction, says: sticker vendors were scarce and sparsely distributed. The “My husband/wife and I are looking for the last week before the data collection, two members of the stickers to complete our kid’s PANINI album. He/She research team mapped the locations of seventy-six street is also looking for the stickers right now, so for the vendors across the five geographic clusters and took pic - moment, I am only quoting. Do you complete lists (of tures of their stalls. missing stickers)?” Based on the initial mapping of vendors, we defined as an inclusion criterion that neighboring vendors must be Given the nature of street-vending in Colombia, quot- at least two meters away from each other (i.e., we chose ing prices without committing to a purchase is an every- them “randomly,” conditional on being at least at a 2  m day interaction. Nonetheless, it is important to discuss distance). We excluded from the sample eighteen ven- two trade-offs created by this script. First, it increases the dors that were too close to each other. Our final sample internal validity. However, it comes at the expense of a includes fifty-nine (78%) of the initially mapped vendors. greater likelihood that sellers could detect that they were Figure  2 shows a detailed description of the vendors’ part of an audit. Second, quoting rather than purchasing location (see Appendix 1). We have substantial variation stickers is useful to leave unaltered the stock of specific in the inclusion rates across clusters. Whereas in the two stickers. Nevertheless, neighboring vendors observing a south-most clusters, the inclusion rate goes above 91%, in failed transaction might update their perception of the the other three clusters, this rate dwells between 53 and buyer’s bargaining skills. We argue that the two caveats of 72%. The inclusion rate is particularly low in the cluster using the script are dampened by the exclusion rule used of Street 147th, where sticker vendors rent temporary to sample vendors and by leaving sufficient time between tents located very close to each other (there are fourteen interactions from the same vendor with two different stalls in 120 linear meters). In our two studies, the inter- research team members. actions with street vendors took place between (Tuesday) Mentioning that the stickers are for the buyer’s kid, June 5th and (Friday) June 8th, one week before the kick- and that these are being simultaneously quoted for both off of the 2018 World Cup. We conducted the experi - parents, helps reducing the commitment for a purchase. ment within four days, on the same week, to minimize Moreover, mentioning the kid on the script aims to pre- substantial variations in the vendors’ stock of stickers. At vent vendors from inferring that men were quoting the most, two buyers were on the field quoting prices on any stickers for themselves, whereas women were doing this particular day. They started to quote lists as early as 9:00 task for someone else. We aim at blocking the latter pos- a.m. and finished as late as 7:30 p.m. One of them visited sibility, which might be interpreted as a confounder of the vendor clusters from South to North of Bogotá, and gender-based discrimination with what  Balafoutas et  al. the other buyer did this process in the opposite direction. (2015) define as “second-order moral hazard.” That is, the tendency of the supply side (i.e., sticker vendors) to anticipate the moral hazard on the demand side (i.e., One week before conducting the experiment, two research team members randomly asked some vendors their usual schedule in each of the sample zones. Given this information, we could guarantee that most vendors would be present when experimenters quoted prices the following week. See Appendix 2 to read the full script. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 5 of 24 6 women quoting stickers for a third party), which results First, 60% of our sample of vendors are close enough to in changes in the price associated with the inference of each other, and therefore looking for these stickers has whom is the final payer. In our design, mentioning the a low transaction cost. Second, vendors typically do not kid makes more likely that male and female buyers are get involved in monetary transactions between them. assumed as final payers of the stickers. Instead, the vendor borrows stickers from another ven- Buyers dressed similarly, with semi-formal clothes to dor to secure the transaction with the buyer, and then represent that they were in their late twenties or mid- gives back the same sticker, or an equivalent one to the thirties, close to their actual age, ranging from 24 to lender. A broad definition of equivalence is defined by the 40 years old. Once the vendor confirms that she/he sells four categories mentioned before: front page emblems, individual stickers, we start reading aloud a handwritten “Legends,” team emblems, and regular stickers. These list of stickers (the set included on the lists is explained “equivalences” also make more plausible the assumption in the next section). The buyer registered every quoted that “mental search costs,” defined as the cognitive costs price of the list, element by element. In case that the ven- of recalling the price of stickers previously sold, are not dor did not have the sticker, we asked for the price she very high either. would charge to find it. In case the vendor said that she The buyer carried a handwritten list and a printed could not find the sticker, we asked for the price she had form. Both included the list of stickers. The latter added previously charged for it in past transactions. This pro - information buyers had to collect prior to the interaction: cedure allowed us to increase the chances of obtaining vendor’s gender, an assessment of her/his age, whether information from every quoted sticker, even if it was not she/he sells other products (related or not to the World available at the time. Cup), type of stall, date and time of the interaction (plus One might think that this procedure makes the audit some empty lines for taking relevant notes regarding the seem “artificial,” as if the buyer put more emphasis on interaction). learning about the prices rather than looking for a transac- Buyers used the handwritten list during the interaction, tion. We provide two reasons why we consider this is not a so if the vendor wanted to grab the list, it would not have major concern. First, from the beginning of the interaction, revealed any additional information about the experi- the buyers mentioned that they were, for the moment, ment. Different members of the research team made the only quoting the stickers. Consequently, since any transac- handwritten lists. After finishing the interaction with the tion would occur another day, asking for the price of cur- vendor, the buyers walked away, hid from the vendor’s rently unavailable stickers would make sense if they could sight, and filled out the printed form. be obtained on a future date. Second, although there are 670 stickers, the total number of price categories is much 2.4 Selection of stickers lower (see Sect. 2.4). Therefore, asking about the price of The FIFA World Cup album’s stickers can be divided into a currently unavailable sticker would have been equiva- two main groups: those printed in shiny foil and those lent to quote another sticker from the same price category printed in regular paper. The latter corresponds to 92.5% (and the latter type of question would have seemed more (620/670) of the entire collection and includes photo- artificial). graphs from players, teams, and stadiums. Most collec- An additional concern about this procedure, in which tors exchange on a 1-to-1 basis these “regular” stickers, we quote unavailable stickers, is that  prices might and most vendors (69%) charge around 1.0 and 2.5 times include search costs. Although this is a valid concern, we the unit cost (i.e., the cost of the packet divided by five, observe that the likelihood that vendors did not have the the number of stickers in the packet) for them. Tables  1 requested sticker but mentioned a price for getting it was and 2 display the list of selected stickers for each study. low. Table  7 (see Appendix 1) reveals that, on average, it We included three “regular” stickers (a stadium, a team, only occurred in 4.35% of the vendor–buyer interactions and a player) in the long lists. The purpose of including (and for 2% of all the potential quoted stickers). Moreo- these stickers was twofold: (i) make more salient the dif- ver, we believe that search costs were low for two reasons. ference between the long list (9 stickers) and the short list (4 stickers) in Study 2, without considerably affecting the difference in total prices, and (ii) make more engaging and realistic the long list for sticker vendors. Vendors recognize four categories of stickers: front page emblems, “Leg- The fifty shiny foil stickers have three main categories ends,” team emblems, and regular stickers (stadiums, teams’ picture, and play- and are significantly more expensive than regular stick - ers). ers. The eight stickers on the front page of the album are The likelihood that vendors did not have the requested sticker and did not mention a price for getting it was also low. Table  7 shows that, on average, the first category. These stickers are related to the World it only occurred in 4.17% of the vendor–buyer interactions (and for 2.1% of Cup (e.g., logo, poster, mascot, ball). The second category the total number of quoted stickers). 6 Page 6 of 24 P. Zamora et al. Table 1 Study 1: Composition of stickers’ lists and variation in buyers’ attributes Experimenter Information ID 1 2 3 4 Nationality Argentinian Argentinian Colombian Colombian Gender Female Male Female Male List quoted Long Long Long Long Sticker type Stickers quoted by each experimenter Local team emblem Colombia Colombia Colombia Colombia Front page emblem 1 FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy Front page emblem 2 Ball Telstar Ball Telstar Ball Telstar Ball Telstar World Champion sticker Maradona (ARG) Maradona (ARG) Maradona (ARG) Maradona (ARG) Team emblem tier 1 Argentina Argentina Argentina Argentina Team emblem tier 2 Uruguay Mexico Peru Uruguay Stadium Samara St. Petersburg Spartak Nizhny Novgorod Team Iceland Denmark Denmark Iceland Player Busquets (ESP) Lallana (ENG) Lallana (ENG) Busquets (ESP) Total Number of stickers quoted 9 9 9 9 on the list Every experimenter quoted one sticker from each row. Most of the stickers are printed on shiny foil, except the stadiums, teams, and players, printed on regular paper. Although we cannot have a formal test to check that the stickers quoted in each category by different buyers are the same, we know that sellers, in general, assign prices based on a limited number of sticker types. Vendors usually do this because having prices for each of the 670 stickers is an enormous cognitive cost for them, and it also increases the menu costs. Additionally, by the very nature of street sales, these are fast transactions, and sellers usually mention prices quickly when one asks for a quotation. Table 2 Study 2: Composition of stickers’ lists and variation in list’s length Experimenter Information ID 3 4 5 6 Nationality Colombian Colombian Colombian Colombian Gender Female Male Female Male List quoted Long Long Short Short Sticker type Stickers quoted by each experimenter Local Team Emblem Colombia Colombia Colombia Colombia Front page Emblem 1 FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy FIFA World Cup Trophy Front page Emblem 2 Ball Telstar Ball Telstar Panini Logo Fair Play Logo Legend Maradona (ARG) Maradona (ARG) x x Team Emblem Tier 1 Argentina Argentina Brazil Germany Team Emblem Tier 2 Peru Uruguay x x Stadium Spartak Nizhny Novgorod x x Team Denmark Iceland x x Player Lallana (ENG) Busquets (ESP) x x Total Number of stickers quoted 9 9 4 4 on the list Every experimenter quoted one sticker from each row. The "x" denotes that this sticker was not quoted in the Short List treatment. The quoted stickers are printed on shiny foil, except the stadiums, teams, and players, which are printed on regular paper comprises thirty-two team emblems, one per each par- the quoted prices that the team emblems are the least ticipating team. Finally, ten stickers presented on the last expensive among these three categories. An exception page of the album form the last category. These stick - is Colombia’s team emblem. Given that the audit was ers are called “Legends” and correspond to the existing conducted in Bogotá, vendors tend to charge the high- World Champions and top-scorers. We learned from est price for this sticker. Since the shiny foil stickers are Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 7 of 24 6 more expensive than the regular ones, in particular those of the audited vendors, allowing us to introduce seller’s on the front page, plus Colombia’s emblem, and the “Leg- fixed effects into the econometric specification. With end,” they open a broader space for price discrimination, this caution in mind, we can check gender differences by either gender-based or nationality-based. comparing prices given to the experimenters with IDs 1 The remaining six stickers included in Study 1 are and 3 vs. 2 and 4; and we can check differences by nation - printed in shiny foil. Two of them are stickers from the alities by comparing prices given to experimenters with front page, one is a “Legend”, and the other three are team IDs 1 and 2 vs. 3 and 4 (see Table 1). emblems. Given the buyers’ nationality, we decided to incorporate Colombia’s emblem and Argentina’s emblem 2.6 Study 2: Price discrimination based on sticker list’s (team in Tier 1). The third emblem is a Latin American attributes team in Tier 2 (either Mexico, Peru, or Uruguay), which The purpose of collectors of the FIFA World Cup album serves as a baseline cost for team emblems. is to complete the whole set of stickers. We argue that, In Study 2, our primary purpose is to detect price dif- when there are few stickers left to complete the collec- ferences per sticker when they belong to a long or a tion, it is plausible that each additional sticker’s marginal short list. In the audit experiment, the long list is identi- value is increasing. In other words, the sticker is valuable cal to the one from Study 1 (indeed, observations from per se, and because collectors derive utility from getting two buyers overlap between studies). For the short list, closer to the completion of the set. Thus, we can test experimenters quoted two stickers from the front page, whether street vendors adopt a sophisticated pattern of Colombia’s emblem, and a Tier 1 emblem: Brazil or Ger- price discrimination, in which they extract a signal of the many. In this sense, the short list included three stickers buyers’ increasing valuation of the missing stickers based where price discrimination might be more pronounced, on the length of the list. We conducted a second experi- plus the Tier 1 team’s emblem as a baseline within the ment in which we hold constant the buyers’ nationality short list. (i.e., all of them are Colombian) to focus on the length of the quoted list. We devised a short list of stickers, com- parable to a subset of what we refer to as the long list 2.5 S tudy 1: Price discrimination based on buyers’ employed in Study 1. attributes Table  2 compares the long list for experimenters with Four experimenters were acting as (potential) buyers. IDs 3 and 4 (i.e., the same data collected in Study 1) with They differed in nationality (two Colombians and two the short list for the new experimenters with IDs 5 and Argentinians) and gender (two males and two females). 6. To test whether the vendors charge more per sticker We expected that vendors could infer that foreigners when the inferred marginal valuation of stickers is higher, were Argentinians due to their pronounced accent; and we compare the quoted prices on the Long List treatment due to some slight variations in the script. For example, (IDs 3 and 4) to the quoted prices on the Short List treat- the script for the Argentinian buyers included the under- ment (IDs 5 and 6). lined words: I am missing one legend, number 664, the one with 3 Data the best player in history: Maradona. The guy is a We collected information on interactions between 59 genius! How much is it? street vendors and 6 experimenters or buyers. We thus Although these additional words make the script have a total of (59 × 6) 354 potential vendor–buyer inter- slightly different for Argentinians, we argue that the actions. We gathered 287 (81%) of such interactions, 199 benefit (i.e., remarking the foreigner buyer’s nationality) for the Long List, in which we had four buyers; and 88 for exceeds the cost (i.e., the risk that these words, and not the Short List, in which we had two buyers. The missing nationality per se, drive the price differences). interactions correspond to cases where the experimenter We exploit those differences to detect price variations could not find a specific vendor. Figure  3 in Appendix 1 between nationalities and between genders. Table  1 maps the interactions of each buyer with vendors. The reports the list of stickers that experimenters quoted in average number of interactions per vendor was 4.86, Study 1. We present the nine quoted stickers in descend- and the median was 5. Forty-one percent of the vendors ing order according to their average quoted price. interacted with all the experimenters, whereas fourteen We acknowledge the limits of our study’s “experimen- percent of the vendors interacted with less than four tal” character since neither gender nor nationality of the experimenters. Taking “sticker × interaction” as the unit buyers were randomly assigned (Siegelman and Heck- of observation, we have a total of 1712 observations in the man 1993). Nonetheless, our design aims to overcome Long List treatment corresponding to four experiment- this limitation by having multiple interactions for each ers and 302 in the Short List treatment corresponding 6 Page 8 of 24 P. Zamora et al. to two experimenters. Note that 1712 is lower than the tend to have more interactions during the afternoon (i.e., (199 × 9 =) 1791 potential observations in the Long List, after 12:00  m) compared to female vendors. We control and 302 is lower than the (88 × 4 =) 352 potential obser- for these variables in our regression analysis. vations in the Short List. The reason is that some vendors did not have some stickers or refused to quote the prices 4 Results individually and instead provided the price of the entire 4.1 Study 1: Econometric results for discrimination based list. on buyer’s attributes Regarding vendors’ characteristics, we find that 42% We estimate a simple linear regression model with robust were in their stall without any companion. According to standard errors and vendor fixed effects to test whether our measure of perceived age, 61% of sellers were 35 years street vendors quote different prices based on the buyer’s old or younger. Concerning vendors’ location, we find nationality or gender. The estimated model is given by: that 43% of vendors are far from any other vendor (more ′ ′ Price = β Foreign + β Male + X α + T δ + ǫ sij 1 j 2 j sij i sij than 50  m away), 44% are at a prudent distance from at (1) least another vendor (between 10 and 50  m away), and the remaining 14% are all located very close to each other The dependent variable is the price of sticker s quoted by (between 2 and 10  m away). As mentioned before, we vendor i to buyer j, expressed in thousands of Colombian excluded from the sample any vendor at a closer distance. pesos [kCOP]. The indicator Foreign equals one if the buyer The right panel of Figure  2 in Appendix 1 shows each is Argentinian and equals zero if the buyer is Colombian. vendors’ location within their geographic cluster. Ven- Similarly, Male equals one if the buyer is a man and equals dors were particularly close to each other in the clusters zero otherwise. In this equation, X is a vector of vendors’ located in Street 85th and Street 147th. For the statisti- characteristics, including age, gender, geographic location cal analysis, we defined only two categories of distance to (cluster), and the presence of neighboring sticker vendors. other vendors: being far from (i.e., more than 50 m away) The vector T includes sticker fixed effects, a categorical isj or close to (i.e., less than 50 m away) any other vendor. variable for the day of the week in which the interactions Since most buyers quoted the price of stickers with took place, and a dummy variable for whether the interac- every vendor, we analyze the interactions as a within- tions occurred in the morning or in the afternoon. subjects design. Nonetheless, we check whether vendors’ Results from Table  3 reveal that prices per sticker characteristics are balanced across the experimenters’ quoted to Argentinians are 208 COP higher than those features of interest in Study 1, and across types of lists quoted to Colombians (about 4.8% higher with respect and experimenters’ features in Study 2, for two reasons. to the mean price of the 1712 stickers in the regression). First, we have some degree of attrition because the exper- However, we do not observe a systematic difference in imenters could not find all vendors. Second, according to prices between male and female buyers. Although male our assignment of vendor ID, the experimenters might buyers are quoted 290 COP more per sticker than female have interacted with different people located in the same buyers, standard errors are slightly larger than the coef- stall. ficients, and therefore we cannot reject the absence of Table 8 in Appendix 1 shows the coefficients of four lin - price discrimination based on the buyer’s gender. Models ear probability models in which we take as covariates the 1 and 2 show that the results are similar using classical or vendors’ gender and (perceived) age, whether the vendor robust standard errors. Given the small number of exper- was isolated, and the journey (morning or afternoon). imenters acting as buyers, we cannot cluster the standard None of the covariates are correlated with the buyer’s errors at the “pseudo-randomization” level. nationality. However, in Study 1, vendors perceived as The econometric model includes seller fixed effects. This older are more likely to interact with male (instead of specification allows us to exploit the within-subject nature female) buyers. In Study 2, male vendors in the Short List of our sampling strategy. In other words, the observed dif- ferences in quoted prices based on buyers’ attributes come from (a subset of) sellers quoting different prices to local and foreign buyers. In this case, the difference we obtain is Three vendors out of forty-eight gave the total price for the list of nine more robust than what would have been obtained by com- stickers to the Colombian buyer instead of quoting the prices individually. Seven vendors out of forty gave the total price to the male buyer for the list of paring quoted prices given by sellers that interacted with four stickers and did not mention individual prices. In the case of the female local buyers to prices given by (comparable) sellers that inter- buyer quoting the short list, five out of forty-eight vendors preferred to give acted with foreign buyers (between-subjects design). Fig- the total price for the four stickers instead of mentioning prices individually. ure 1 depicts some coefficients of interest from Eq.  1. Panel The statistics presented in the text correspond to the average for the six experimenters. Since it is possible that buyers interacted with different A shows that vendors quote a higher price for shiny foil stick- vendors at the same stall, we first compute these demographic statistics per ers, particularly for stickers on the album’s front page and experimenter and then we average the reported percentages. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 9 of 24 6 Table 3 Eec ff t of buyer’s attributes (nationality and gender) on One of the advantages of creating a list with multiple street vendors’ price quotations items to quote is the opportunity to explore more sophis- ticated patterns of price discrimination. Since we verified Variables (1) Price [kCOP] (2) Price [kCOP] the existence of price variation across sticker types (see Foreigner 0.208* (0.113) 0.208* (0.120) Panel A in Fig. 1), we now conduct the regression analysis Male Buyer 0.290 (0.303) 0.290 (0.369) separately for each sticker on the list. Constant − 0.168 (0.326) − 0.168 (0.284) Table 4 reveals the results for the price quoted for each R-squared 0.799 0.799 sticker. Additionally, Column 1 shows the price quoted Controls and Yes Yes for the full list. This result confirms that vendors charge Vendor FE larger prices to Argentinians than to Colombians (on Errors Standard Robust average, 2,438 COP more). Nonetheless, this pattern Vendors 59 59 is not uniform across stickers: the second most expen- Number of 1712 1712 sive sticker (i.e., the FIFA Trophy), and the two cheapest observa- stickers on the list, drive the higher prices quoted to for- tions eigners. The coefficients for the two other most expensive *p < 0.1, **p < 0.05, ***p < 0.01. Standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in stickers, Colombia’s emblem and the Official ball-Telstar, thousands of Colombian pesos (kCOP). The mean of the dependent variable show a pattern similar to the FIFA Trophy, though they is 4.275 kCOP. Controls in all regressions: geographic cluster, day and sticker are less precisely estimated. fixed effects; (perceived) age and gender of the vendor, whether the vendor has nearby vendors, and time of the day We offer a conjecture for this pattern, in which Argen - a Stickers Colombia's FIFA World Ball Maradona Argentina's Tier 2 Team Stadium Team Player Emblem Cup Trophy Telstar (Legend - ARG) Emblem Emblem b Zones Downtown Street 72 Street 85 Street 93 Street 147 Fig. 1 Coefficients for additional covariates in the regression shown in Table 3: (A) sticker fixed effects, (B) geographic clusters. The color intensity of confidence intervals represents the confidence level (from darker to lighter, 90, 95, and 99%). The reference category in Panel A is the sticker of a player (i.e., the cheapest sticker). The reference category in Panel B is downtown (the area of Bogotá in which stickers are the cheapest) Colombia’s emblem. Panel B reveals that stickers’ price dif- tinians are simultaneously charged more for the most and fers between spatial clusters, increasing as we move to the the least expensive stickers. On the one hand, vendors Northern side of Bogotá. have more discretionary power to raise prices of the most expensive stickers, arguing that they are in high demand. An explanation for the considerably larger price in the farthest cluster One could interpret this result based on the heteroge- (Street 147th) is that sticker vendors rent temporary tents located in a small mall’s borders at a significant price (about 1.5 monthly minimum wages). It neity in price discrimination detection (Grossman and leads to a very high concentration of vendors (fourteen in 120 linear meters) Honig 2017). This pattern is not observed for other shiny that, despite their spatial closeness, manage to keep the city’ highest prices. foil stickers with a small variation in their price, such as However, given their spatial proximity and their long-lasting reputation of selling stickers in every World Cup, it is also possible that vendors in this area team emblems. On the other hand, for teams and players, are more likely to collude. kCOP kCOP 6 Page 10 of 24 P. Zamora et al. ff Table 4 Eect of buyer’s attributes (nationality and gender) on quoted price, per type of sticker (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Full List of nine Colombia’s FIFA World Cup Official ball Maradona Argentina’s Emblem Tier 2 Team Team Player Variables Stadium Stickers Emblem Trophy -Telstar (Legend - ARG) Emblem Foreigner 2.438* (1.305) 0.467 (0.338) 0.680** (0.300) 0.328 (0.284) 0.042 (0.328) − 0.010 (0.191) -0.063 (0.208) 0.108 (0.099) 0.174* (0.092) 0.161* (0.089) Male Buyer 3.200 (6.334) 1.133 (1.450) 0.564 (1.054) 0.192 (1.111) − 0.273 (0.967) − 0.074 (0.655) 0.499 (1.029) − 0.050 (0.233) 0.087 (0.206) 0.224* (0.123) Constant 33.941*** (4.856) 3.920*** (0.969) 6.254*** (1.121) 5.968*** (1.170) 6.225*** (0.883) 4.442*** (0.589) 4.235*** (0.685) 0.747*** (0.212) 0.750*** (0.189) 0.285*** (0.099) R− squared 0.766 0.817 0.749 0.716 0.668 0.699 0.596 0.736 0.715 0.522 Controls Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes and Ven- dor FE Vendors 59 59 59 59 59 59 59 59 59 59 Number of 187 193 187 187 188 192 192 190 191 192 observa- tions *p < 0.1, **p < 0.05, ***p < 0.01. Robust standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in thousands of Colombian pesos (kCOP). Controls in all regressions: geographic cluster, day and sticker fixed effects; (perceived) age and gender of the vendor, whether the vendor has nearby vendors, and journey. In Column (1), we use the list’s total price if the vendor mentioned it. Otherwise, we use the sum of the nine stickers for each vendor–buyer interaction when the vendor separately quoted the prices for the nine stickers. We could not construct the sum in 12 interactions (10 vendors) because the vendor did not quote some stickers. In 9 interactions (8 vendors), the vendors gave a total price for the nine stickers. Four of them did not say the price for each sticker individually, and five of them proposed to give the Stadium, the Team, and the Player for free if the buyer bought the other six stickers. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 11 of 24 6 the least expensive stickers on the list, the modal prices for each of the listed (and comparable) stickers. We do are 1000 and 500 COP, respectively. That is, they are not find evidence suggesting that vendors charge a larger between five and ten times cheaper than the least expen - price for the same sticker when it belongs to a short list sive shiny foil stickers. Figure 4 reveals that foreigners are compared to when it belongs to a long list. Although the more likely to be charged prices two or three times larger number of observations is smaller in this econometric than the modal price for these regular stickers. Another exercise (the sticker list is shorter and Argentinians were mechanism that might be at play for the least expensive excluded), coefficients for the indicator variable for the stickers emerges from the fact that regular stickers are Long List variable are very close to zero. In other words, sometimes offered “as a gift” when purchasing the shiny the lack of significance does not reflect a lack of statistical foil stickers. However, we do not find evidence that these power. gifts are more likely to be offered to Colombians. We have two conjectures for the absence of an effect. Regarding the analysis of gender differences, Table  4 The first one is that sellers do not exploit the (assumed) provides additional, suggestive evidence that male buy- higher marginal valuation for fewer missing stickers. ers do not appear to receive systematically different The second conjecture is that, although one of the lists is quoted prices compared to female buyers. Nonetheless, twice as long as the other, both lists could be considered these results must be taken with caution since the sam- by the vendor to be “short,” in absolute terms. ple size sharply decreases at the sticker level. Similar to the discrimination against Argentinians, it might be the case that men are also charged higher amounts for the 5 Qualitative results from interviews most expensive stickers, particularly for the Colombian In addition to our audit studies, we conducted eleven emblem. interviews with sticker vendors that were part of our As an additional check, we add to the econometric original sample. At least two vendors were inter- model the interaction between the buyer’s and the ven- viewed per geographic cluster. The purpose was to dor’s gender. Table 9 reveals that gender interaction vari- complement our insights from this market and elicit ables are not statistically significant. Therefore, even if sellers’ perceptions of price discrimination. The we consider more specific mechanisms in which male interviews were conducted four weeks after the audit vendors might discriminate against male buyers but not experiment. By that time, the knockout-stage of the against female buyers or vice versa, we do not detect FIFA World Cup was already taking place, and the sale price differences based on gender. of stickers was less intense, so it was feasible to ask sellers to participate in our interview. The interviews were conducted on July 2, 2018, an official holiday in 4.2 S tudy 2: Econometric results for discrimination based Colombia, for the two clusters corresponding to resi- on list’s attributes dential areas. We selected this day because there were We discussed earlier that stickers might have an increas- substantially fewer people, and the probability of the ing marginal utility when the buyer’s purpose is to vendor accepting the interview was higher than on a complete the stickers collection. Vendors could then business day. The other three clusters correspond to anticipate a greater willingness to pay for the same sticker economic activity centers, and sticker vendors do not from a buyer revealing a short list of missing stickers, work on holidays. Hence, we conducted the interviews with respect to a buyer revealing a long list of miss- the day after. ing stickers. We conduct a regression similar to Eq.  (1), We opted for conducting eleven interviews due to replacing the variable that captures buyer’s nationality budget constraints since we offered compensation for (since all the experimenters in Study 2 are Colombians) the vendors’ time. We selected the vendors based on for a variable indicating whether the quotation corre- three attributes. First, vendors for whom we had com- sponded to a short or a long list. We focus here on the pleted the interactions with each experimenter (six in quoted price for the four stickers listed in Table  2 that total). The reason was to maximize the chances of find - are common to both lists: Colombia’s emblem, the FIFA ing the selected vendor by the time of the interview. Sec- World Cup Trophy, an additional shiny foil sticker from ond, since we had six observations per selected vendor, the front page (i.e., the Official ball-Telstar, the FIFA Fair we aimed at guaranteeing that it was the same person in Play logo or the Panini Logo) and the emblem from a Tier 1 team (Argentina, Brazil or Germany). We report the comparison between the Short List and One final interview was completed ten days later (July 13). We decided the Long List treatments in Table 5. The first column cor - to make this last interview to obtain an additional testimonial from the area responds to the model pooling all four stickers, whereas where vendors were closer to each other but from one of the few isolated ven- the next four columns display the coefficients of interest dors in this area. 6 Page 12 of 24 P. Zamora et al. Table 5 Eec ff t of sticker list’s attribute (length) on quoted price, per type of sticker Variables (1) Full list of four (2) Colombia’s Emblem (3) Front Page Emblem (4) Front Page Emblem (5) Tier 1 Team stickers 1 2 Emblem Long List 0.056 (0.113) 0.095 (0.237) 0.021 (0.150) 0.046 (0.155) 0.014 (0.127) Male Experimenter -0.029 (0.122) 0.221 (0.247) -0.204 (0.147) -0.261* (0.150) 0.013 (0.126) Constant 4.064*** (0.438) 3.034*** (0.714) 4.433*** (0.458) 4.322*** (0.424) 3.650*** (0.431) R-squared 0.553 0.803 0.838 0.811 0.709 Controls and Vendor FE Yes Yes Yes Yes Yes Number of observa- 697 176 172 173 176 tions *p < 0.1, **p < 0.05, ***p < 0.01. Robust standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in thousands of Colombian pesos (kCOP). Controls in all regressions: geographic cluster, day and sticker fixed effects; (perceived) age and gender of the vendor, whether the vendor has nearby vendors, and journey. Front Page Emblem 1: FIFA World Cup Trophy. Front Page Emblem 2: Official ball—Telstar, FIFA Fair Play Logo, or Panini Logo. Tier 1 Team Emblem: Argentina, Brazil, or Germany. In Column (1), we use the list’s total price if the vendor mentioned it. Otherwise, we use the sum of the four stickers for each vendor–buyer interaction when the vendor separately quoted the prices for the four stickers. We could not construct the sum in 3 interactions (3 vendors) because the vendor did not quote one of the four stickers. In 24 interactions (21 vendors), the vendors gave a total price for the four stickers. Eleven of them did not say the price for each sticker individually. all the interactions. Third, we guarantee one man and one respectively. The remaining one did not answer. Three woman were interviewed in each cluster. vendors declared to contribute to the pension system or The interviewer approached the vendor and asked if health system. she/he could respond to some non-personal questions Two interviewees in Street 147th were located in that would take between 15 and 20 min. We list the ques- fixed stalls. One paid a monthly rent of approximately tions in Appendix 1. 1,200,000 COP, and the other paid 1,700,000 COP (between 1.5 and 2 monthly minimum wages). The rest of the interviewees had a mobile stall. Although they could A. The interviewer would buy one of the shiny foil stick - change locations within or between days, sticker vendors ers for 10,000 COP in appreciation. By the time of the tend to stay in a fixed spot during the World Cup. All but interview, the price of such stickers was about 4,000 one of the interviewed vendors has stayed in the same to 6,000 COP. All the vendors initially approached spot since at least the 2010 FIFA World Cup. Despite by the interviewer agreed to participate. We did not the Constitutional protection of street vendors, they are mention any relationship between the interview and sometimes subject to police inspections that might lead the audit study, as it may have caused unpleasantness to confiscation of their merchandise. In these cases, ven - among the interviewees. We requested permission to dors with mobile stalls locate nearby their usual spot. record the interview, and all of them accepted. 5.1 Pricing strategies We interviewed five women and six men with the We identify three different types of strategies based on following characteristics. The interviewer perceived the responses from eight interviewees, which we define eight of them as being older than 35  years. Five of them as frequency-based, distributor-based, and imitation. declared to work in their stall alone, whereas the other The most usual strategy is what we define as frequency- six declared to work with at least one more person. In based. It consists of opening a vast number of boxes, each most of the cases, this companion person was not a rela- containing 104 packets with five stickers each, for a total tive. Working hours from interviewed vendors went from of 520 stickers per box, and then inferring each sticker’s 8 to 12.5 h a day, with an average of 10.4. Depending on rarity to assign prices based on how scarce they are. The the geographic cluster, they work on weekdays (Down- following is an example of a vendor using this strategy: town and 72nd Street) or the whole week (85th, 93rd, and 147th Street). Some of them claimed that they were [Excerpt 1] “Well, I already have experience with working more hours than usual given the World Cup the World Cup, right? I open boxes because you must Season. Eight of the interviewees were also dedicated open boxes or master boxes as people say… because to street vending the rest of the year, and they sell other seasonal products (e.g., Christmas merchandise) or spe- cific products such as flowers, fast food, umbrellas, and The responses of the three remaining vendors cannot be linked to any strat - used books. Another two vendors who replied to this egy. These were: “based on the price of the box”, “shiny foil stickers cost 4,000 COP, except #1 that is more costly, and any other sticker costs 300 COP”, and question worked in tailoring and building maintenance, “this is my fourth World Cup selling stickers, so based on my experience.” Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 13 of 24 6 a master box contains 12 boxes. So let’s say I open 3, average prices are also higher than in the other clusters. It 4, master boxes. Then, by arranging the stickers by is also worth mentioning that, whereas they perceive the numbers, I deduce the team that comes out the most, information provided by other vendors as useful, several the players who come out the most, and the stickers vendors also claimed that they dislike when buyers tell that come out the least. Hence, those that come out them that a nearby vendor is selling the same sticker for a the most are the cheapest, and those that come out lower price. Regardless of whether the prices provided by the least are the most expensive.” buyers were truthful or were part of a bargaining strategy, vendors appeared to dismiss this information. When asked about the number of opened boxes, the responses ranged from 24 to 100. This process gives 5.2 Awareness of price discrimination between 12,500 and 52,000 stickers, which will be suf- We included four questions regarding perceived price ficient to complete between 18 and 77 collections in the variations. First, we asked vendors whether they changed absence of duplicate stickers. The frequency-based strat - the prices of stickers within a given day. Ten of them egy requires a substantial investment since each box ini- responded negatively. The only vendor who provided tially cost them around 180,000 COP. Opening 24 boxes a different response claimed that he would change the would cost 4,320,000 COP, more than five times the price of stickers within a day if the number of units left monthly minimum wage at the time. of a specific sticker was too low. This vendor followed a The distributor-based strategy consists of following distributor-based pricing strategy. the prices per sticker suggested by their PANINI dis- We also asked vendors whether they had changed the tributor for most of the stickers. In general, vendors were prices of stickers between days. Ten of them responded instructed to charge 300 COP per sticker and comply affirmatively, arguing that prices had been declining over with most stickers. However, they tended to charge more time since mid-March when the distribution of stick- for the shiny foil stickers and others that they considered ers and albums started in Bogotá (the remaining vendor rare. responded negatively). Even though most of them agreed Finally, imitation consists of adjusting to the prices set on the price decline, they provided different explanations by other vendors. They gather information on sticker for this trend. Five of them argued that the reason for the prices by directly asking neighboring vendors. For downward trend was that the price of the box contain- instance, a vendor said: ing 520 stickers had reduced by approximately 30%. Two [Excerpt 2] “We set prices based on past sales expe- vendors claimed that prices lowered as a consequence rience and checking with other sellers.” of greater competition given the large number of ven- dors, whereas two other vendors claimed that demand One of the interviewed sellers claimed that sometimes had been decreasing. The remaining vendor claimed that they have to lower prices even if it leads to losses because price fluctuation was mostly associated with the perfor - their neighboring vendors create much competition. mance of the Colombian squad in the World Cup. More precisely, she says: We introduced first these two questions in which vendors would be more likely to acknowledge price variation than [Excerpt 3] “Well, I have tried to change the prices, when being asked about price changes that may hint discrim- because at least with the stickers, the shiny ones are ination based on buyers’ attributes. Then, we proceed to ask if the difficult ones and then I am the one that sets they charged different prices to males and females, and locals the highest prices. The rest of the stickers are cheap with respect to foreigners. All the respondents said that they because of the price war. It is about the one who sells charged the same prices regardless of gender. Similarly, all but more… is not the one who earns the most, but the one of the responders said that they charged the same prices one who sells the most… is the one who wants to fight regardless of nationality. The remaining vendor said: for a client, the one who gives him/her the lowest price. Also, clients put a price on the merchandise. [Excerpt 4] “Not really… the price is like the same. In other words, you cannot set the price because they What changes is as people say… as the saying goes… already come to impose the prices.” to the highest bidder, right?… then let’s say, those who are charged a little more, or something like that, The two vendors following an imitation strategy were maybe are the foreigners.” located on the cluster in the 147th Street, where vendors’ stalls are closer to each other than any other cluster, but The audit experiment and the interviews provide dif - ferent responses to whether sticker vendors discriminate against foreigners. The main challenge for interpreting the The official price per box to the public was 218,400 COP, about 21% more interviews’ responses is to disentangle between a social expensive. 6 Page 14 of 24 P. Zamora et al. desirability bias (i.e., vendors might believe that acknowl- geographic clusters, from 25,000 to 40,000 COP. Vendors edging price discrimination is objectionable) from a pat- also claimed that distributors started to give them similar tern of implicit (or unconscious) discrimination (Bertrand large sheets with multiple shiny foil stickers when they et al. 2005). An argument against the social desirability bias bought boxes containing packets of stickers. According to is that sellers acknowledge the trade of counterfeit stickers. a vendor, this was a compensation offered by the distribu - This argument is explored in-depth in the next section. tors, given the rarity of shiny foil stickers. Seven of the interviewed vendors claimed that they 5.3 Counterfeit stickers were aware of counterfeits, but they were not selling The rarity of shiny foil stickers in the 2018 World Cup them. The main reason exposed to not engage in the sale album gave origin to the presence of counterfeit cop- of counterfeits were reputational concerns. In most cases, ies among street vendors, as some of them confirmed. they do not want to lose their clientele since they have Regarding this rarity, one vendor said: been selling stickers in the same spot for at least three FIFA World Cups. Two of the vendors revealed that, [Excerpt 5] “In the last World Cup, there were sev- although they were not selling counterfeits, sometimes eral teams that came out a lot and others that did they exchanged stickers with buyers and, as a result of not. However, this year all the teams came out, as these exchanges, they ended up with counterfeits. The well as all the players. What did not come out were following is the testimonial from a vendor who acciden- the shiny ones. Let’s say, from a single box, there tally received counterfeits in an exchange: were 18, 19 shiny stickers coming out. There are 50 in total, considering the first and last page, as well [Excerpt 7] “Since I have the stall, I must sell what is as the emblems. Usually, 18 came out from a box, at original, but on the street, there are a lot of scanned most 20 or 22, you see?”. stickers that vendors are selling, and there are peo- ple who do not know. Then what happens? one day As a consequence of this rarity of shiny foil stickers, I had a problem with a DIAN official. He was a vendors confirm that people passed by offering them good client, and he brought colleagues here. It turns a set of 50 shiny foil counterfeit stickers. For instance, a out that I, by mistake, for being there (points out vendor said: to the street), and in the bookstore, I received some [Excerpt 6] “They pass by asking: < < Do you need counterfeit stickers, I exchanged them, and then I shiny stickers? > > , and I answer : < < Yes, but origi- was going to sell them. It turns out that precisely I nals > > , and then they say: < < Nooo, you know that gave them to him! and the official brought me the I don’t have those > > , so I respond: < < Ahm no! I police!… but I did not know, I did not know that the already have those, I need originals. > > They sell the stickers were fake. They are very well made, but nev - set of 50 shiny stickers for 25 (thousand COP) or 20 ertheless, there is a detail that you can recognize.” (thousand COP).” The remaining four vendors acknowledged that they were selling counterfeits. However, they sold both types Learning about the existence of counterfeit stickers was of stickers and claimed that they offered the originals one of our major motivations to conduct the interviews first. Only when buyers said the stickers were too expen - since price variation could reflect possession of counterfeits sive, they did offer counterfeits as a cheaper alterna - rather than price discrimination based on buyers’ attributes, tive. The following is the testimonial from one of these or even worse, a combination of both. Think, for instance, vendors: that the same price was quoted to locals and foreigners, but vendors offered original stickers to the former and counter - [Excerpt 8] “Look, I’m very honest with the client. I feits to the latter, then the two prices would not be compa- didn’t use to bring fake stickers, but clients asked me rable. We argue that this was not the case since counterfeits for fake ones… and now I bring fake ones. So I tell the appeared in the market, but they were only offered after client: < < Well, do you want cheap or do you want buyers claimed that shiny foil stickers were too expensive. expensive? Do you want the original or not? > > … Three different vendors mentioned that they started to be offered a large sheet containing the fifty shiny foil stickers in the collection two weeks before the kick-off A street vendor located in downtown said that the price of these 50 fake of the World Cup. The price of this sheet varied between stickers also fell over time like the original stickers:”they sell the package at 25,000, or 20,000. In the beginning, they sold it at 50,000.” With terms of trade highly advantageous for vendors, such as 3 for 1 sticker. However, buyers regularly accepted because they were close to com- plete the collection and had a large stock of duplicates. DIAN is the Colombian Tax and Customs National Authority. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 15 of 24 6 then, it depends on the customer. I tell them: And another one mentioned how investments were < < well, you are consciously buying a fake one > > . ” driven by expectations about the performance of the National squad four years before: The presence of counterfeits is not very likely to affect the outcomes in our audit experiment for two reasons. [Excerpt 10] “Four years ago, the jerseys were scarce. First, according to a couple of vendors, counterfeits were Colombia won, and the price went up… it immedi- easily detectable based on the lower brightness and the ately went up. So I said, this year, the same thing will lack of holograms, and it is therefore credible that they happen, and it was not true. I received a loan of 10 were not offered to buyers in our audit study. Second, our million pesos, and I invested it in a lot of jerseys. I script did not allow buyers to bargain over prices. There - paid 23,000 pesos (each), and today I went to buy fore, the typical offer of counterfeits as a response to a some small jackets for kids, and I realized that the price reduction was not likely to occur. (jerseys’) price now is 4,000 pesos… Too bad!… Now I can cry over spilled milk. If they win today, I go and buy. From there, I recover some of what I… It’s ter- 5.4 O ther insights from the stickers market rible, terrible. Do you know what that is? Look, we This last subsection is devoted to the analysis of three bought the jerseys for 23,000 pesos, and we are sell- additional lessons on how the stickers market works. ing them for 25,000. Can you imagine what it is like Some of these insights yield more general lessons regard- waiting for four years to make a profit of 2,000 from ing seasonal street-vending, primarily linked to sporting a jersey (0.7 US Dollars)?”. events. This testimonial also confirms the large investments in From the audit experiment and the interviews, we sporting events merchandise made by street vendors. Our learned that the stickers market is subject to multiple second insight is that these loans, combined with the long sources of uncertainty. The spatial distribution of ven - periods that street vendors have stayed in this business, dors, together with the low relocation costs in some suggest that (in this seasonal market) informality is a choice zones, revealed strategic uncertainty in the competition rather than an outcome associated with exclusion of the for clients. Additionally, the vendors also disclosed they formal labor force (Levy 2010; La Porta and Shleifer 2014). had made large investments and were concerned by the The ten interviewees that responded for how long they had rapid drop in prices of the merchandise associated with been street vendors during the World Cup season men- the World Cup. In fact, they mentioned that they were tioned on average 14.9  years. Nonetheless, some of them forced to decrease prices sooner than usual due to the have other off-season jobs different from street-vending. large competition that year and the number of counter- They take advantage of the flexibility from informal labor feit stickers circulating in the market. markets to work on the streets when it is more profitable. Another source of uncertainty we could identify from A final insight is how the costs of entry and exit from sea - the interviews was the connection between vendors’ sonal street-vending relate to the spatial location of vendors. profits and expectations and the performance of the In principle, if entry is not costly, vendors’ location is subject Colombian national squad in the World Cup. This link is to the “tragedy of the urban commons” (Webster 2007; Kim important since vendors invested in merchandise based 2012). Two reasons alleviate this problem. On the one hand, on the outcome of the Colombian national squad in the attractive selling spots are less scarce given the sparse loca- previous World Cup, which was exceptional. In addition, tion pattern nearby stores of a high-end supermarket chain. the performance during the 2018 FIFA World Cup not Besides, most street vendors do not see each other as fierce only shaped expectations about sales but also determined competitors. This is the testimonial of a vendor when asked the price of the merchandise, particularly of stickers. about her location nearby one of these stores: Vendors were afraid of an early elimination of the Colom- bian squad, which would drastically decrease the demand [Excerpt 11] “My spot was next to the door of Banco for goods related to the World Cup. The converse appears de Bogotá, and then I got sick and went home. There, to be also true for some vendors. For instance, a vendor on the sidewalk, there was a thing on the (parking) replied the following when asked about adjusting the blue zones, and I had a stall there, but then since I price of stickers: got cancer, I went home. Now I come here to work only during the World Cups [Interviewer: Ok… [Excerpt 9] “It depends… At least now during the and is it better to be around other vendors like this? World Cup, as long as Colombia wins, (prices) go Because generally there is a lot of people here] A lot! up.” Look, let me tell you one thing: what God has for you, he keeps it, no matter where you are.” 6 Page 16 of 24 P. Zamora et al. On the other hand, there are some areas where the rela- We combined the audit studies with eleven interviews tionship with policemen mediates location outcomes. In conducted with sticker vendors from the audited sample. most cases, entry costs might be higher since policemen ask The interviews shed light on crucial aspects to interpret the for merchandise (i.e., stickers or jerseys) to let them work on results from the audit studies. Interviewed vendors claim those areas. However, policemen might also give priority to that they do not discriminate based on buyer’s attributes. vendors established in a given spot for a  long time, under We argue that these responses might reflect implicit dis - some conditions. A vendor revealed that she has to work, crimination (Bertrand et al. 2005) rather than a social desir- once a week, on a spot where sales are good only in Decem- ability bias. Vendors concerned about providing socially ber (otherwise policemen might reassign her spot). She also desirable answers would have not naturally claimed that revealed that policemen transfer the costs of reducing occu- they sold counterfeits in addition to official stickers, as pation of the public space to vendors. This is her response: occurred with about half of them. From a more general perspective, the combination of audit studies with inter- [Excerpt 12] “Yes, (they let us work) under the condi- views allowed us to better understand the functioning of tion that we do not let any other new (vendor) get in” . this seasonal market. Although street vendors make large investments in merchandise related to the sporting event (including stickers), price variation is much better explained by the vendors’ location around the city and not by altering 6 Final discussion prices to specific buyers in a considerable amount. Street-vending of stickers is an informal market without The recent use of audit experiments in developing posted prices, which allows sellers to exercise a discre- countries introduces detectability as an important ele- tionary power. We conducted two audit studies on sticker ment for understanding behavior. For instance, Gross- vendors to test the existence of price discrimination in man and Honig (2017) show that rice sellers are more this context. Study 1 aimed at detecting price discrimi- likely to discriminate against High-class non-coethnics, nation based on two buyers’ attributes: nationality and not by charging directly a higher price, but by deliver- gender. We find evidence of price discrimination against ing smaller quantities, conditional on paying the same Argentinians with respect to the local Colombian experi- amount for a scoop of rice. Similarly, we found that price menters. Foreigners were charged on average 4.8% more discrimination based on nationality was most evident for per sticker, and the effects are significant for the most the expensive stickers, those in which price fluctuations expensive and the least expensive stickers. We argue that and scarcity can be used to argue discretionary prices. sellers use the higher price fluctuation of the most expen - This evidence calls for further studies focused on the sive ones, for which they can claim scarcity, as an excuse detectability of discrimination in informal markets. to increase prices for foreigners. For the least expensive Open questions include whether the subtleness of stickers, sellers infer that foreigners are less familiar with discrimination operates at a conscious or unconscious the low prices of regular stickers (i.e., those not printed in level, and whether the lower detection of discrimina- shiny foil), and is more likely that they receive quotations tion emerges in developing countries as a response to that are two or three times higher than the modal value. the fear of retaliation from the buyer. When formal By contrast, we did not find evidence of gender dis - and informal markets coexist, this question becomes crimination. Our script was designed to make sure that even more relevant since sanctioning mechanisms dif- any price difference between male and female buyers was fer between markets. If such fear of retaliation exists, not driven by a problem of second-order moral hazard it would be interesting to understand whether the per- (Balafoutas et al. 2015), in which vendors infer that men ceived threat is associated to direct punishment (i.e., quoted stickers for themselves, whereas women quoted verbal or physical aggression, or reports to the police) stickers for someone else. or indirect punishment (i.e., abstain from trade) (Bala- Study 2 aimed at detecting price discrimination based on foutas et al. 2014). Finally, fear of retaliation can lead to a good’s attribute: the length of the quoted sticker list. We models of multidimensional statistical discrimination, hypothesized that vendors could have inferred that buyers in which buyers’ (resp. vendors) attributes not only sig- having fewer stickers left to complete the collection would nal a willingness to pay (resp. to accept), but also indi- have a higher willingness to pay, which would have encour- cate the likelihood and expected damage of retaliation. aged them to raise the price of the same sticker when it belonged to a shorter list. We do not find evidence of this more sophisticated discriminatory behavior. Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 17 of 24 6 Appendices Appendix 1 See Tables 6, 7, 8, 9 and Figures 2, 3, 4 We present three figures in this appendix. The first Appendix 2 two figures illustrate vendors’ location and their interac - Script for experimenters acting as vendors tion with experimenters acting as buyers. The left side of In the following scripts, squared brackets correspond to Fig.  2 depicts a map of the Northeastern side of Bogotá, instructions for the buyer that must not be read aloud. including the five clusters of sticker vendors. The right Sticker numbers separated by a slash (/) correspond to side of Figure 2 zooms in each one of the clusters to pro- the randomization of similar stickers between lists. The vide better information on vendors’ spatial location, text wrapped in < < > > is mentioned only by Argentinian Figure 3 shows a grid of buyers in the columns and ven- buyers. dors in the rows. The dots and diamonds indicate each vendor–buyer interaction registered in our experiment. Script #1: Long list [Before starting: Take out both the Figure  4 shows the cumulative distribution of quoted handwritten list and the printed list from the envelope. prices for the three least expensive stickers (i.e., those Check that the vendor’s ID matches the letters in the printed on regular paper), separately for Colombians and envelope. Mark all the information that is on the back of Argentinians. Table 6 Day on which each experimenter quoted prices and the route that followed Day Tueday June 5 Wednesday June 6 Thursday June 7 Friday June 8 Experimenter / Route COL-F-LL / North to COL-M-LL / South to North COL-F-SL / North to South ARG-F- South LL / North to South Experimenter Route ARG-M-LL / North to South COL-M-SL / South to North COL-F-LL refers to the Colombian woman in the Long List treatment. The COL-M-LL experimenter went back to Downtown during the morning on Friday, June 8, since there were scarce vendors during the morning on Wednesday, June 6. There was a large police presence in the area, which caused the vendors to disperse and made it difficult to find them. Also, the ARG-M-LL experimenter went back to Downtown during the morning on Thursday, June 7, since there were scarce vendors during the late afternoon the day before Table 7 Proportion of vendor–buyer interactions with unavailable stickers a Vendor did not have the sticker and did not give any price Vendors Stickers COL-F-LL ARG-M-LL COL-M-SL COL-F-LL ARG-M-LL COL-M-SL Downtown 0/21 1/12 1/6 0/189 6/108 1/24 Street 72nd 0/13 0/10 0/13 0/117 0/90 0/52 Street 85th 0/9 1/9 1/7 0/81 8/81 1/28 Street 93rd 1/7 0/8 1/7 6/63 0/72 1/28 Street 147th 0/8 0/7 0/7 0/72 0/63 0/28 Total 1/58 = 1.72% 2/46 = 4.35% 3/40 = 7.5% 6/522 = 1.15% 14/414 = 3.38% 3/160 = 1.87% b Vendor did not have the sticker but did mention a price Vendors Stickers COL-F-LL ARG-M-LL COL-M-SL COL-F-LL ARG-M-LL COL-M-SL Downtown 0/21 1/12 1/6 0/189 4/108 1/24 Street 72nd 0/13 1/10 2/13 0/117 3/90 3/52 Street 85th 0/9 0/9 0/7 0/81 0/81 0/28 Street 93rd 1/7 1/8 1/7 3/63 6/72 1/28 Street 147th 1/8 0/7 0/7 1/72 0/63 0/28 Total 2/58 = 3.45% 3/46 = 6.52% 4/40 = 10% 4/522 = 0.77% 13/414 = 3.14% 5/160 = 3.12% Columns correspond to the three buyers who collected this information in detail. COL-F-LL refers to the Colombian woman in the Long List treatment. For the Long List, the number of missing values in bold font coincide with the number of missing values in the database for the experimenter. For the Short List, even though nine vendors did have the four stickers, they only mentioned the price for the full list, and not for each of them individually 6 Page 18 of 24 P. Zamora et al. Table 8. Balance across attributes made salient in Studies 1 and 2. Coefficients correspond to linear probability models. (1) (2) (3) (4) Sample Long List Short List Full Sample Dep. Variable Gender (=1 if Male Buyer) Nationality (=1 if Foreigner) Gender (=1 if Male Buyer) List Length (=1 if Long List) (Perceived) Vendor’s age (= 0.232*** (0.072) − 0.040 (0.076) − 0.056 (0.105) 0.024 (0.076) 1 if 35) Vendor’s gender (=1 if 0.032 (0.073) − 0.046 (0.075) − 0.068 (0.106) − 0.018 (0.076) Male) Isolated vendor 0.094 (0.075) 0.008 (0.077) 0.026 (0.115) − 0.084 (0.076) Journey (=1 if Afternoon) − 0.129 (0.092) − 0.010 (0.091) 0.379*** (0.112) 0.170** (0.085) Constant 0.373*** (0.099) 0.522*** (0.101) 0.262*** (0.117) 0.454*** (0.092) Observations 199 199 88 194 R-squared 0.058 0.003 0.129 0.024 *p < .1, **p < .05, ***p < .01. Robust standard errors are shown in parenthesis. The dependent variable in all specifications is the vendor’s price quoted in thousands of Colombian pesos (kCOP). The sample size in model (4) is 194 observations because Foreign buyers were excluded from the regression ask her for the price that she would charge for it if she Table 9 Eec ff t of vendor–buyer gender interactions on the price finds it for you. If the vendor says she could not find it, quoted by the vendor ask for the price she charged the last time she sold it.] Okay, let’s start with the team emblems. I need three. Variables (1) Price [kCOP] (2) Price [kCOP] First of all, I need the emblem of my country: Colom Foreigner 0.203* (0.113) 0.203* (0.121) bia/Argentina. It’s number 620/260. How much is it? Male Buyer x Female Vendor 0.336 (0.312) 0.336 (0.381) I also need the emblem of Argentina/Colombia. It’s Female Buyer x Male Vendor 0.153 (0.166) 0.153 (0.178) number 260/620. How much is it? And the emblem of Male Buyer x Male Vendor − 0.097 (0.155) − 0.097 (0.157) Peru/Uruguay. It’s number 220/80. How much is it? Constant − 0.200 (0.330) − 0.200 (0.288) I am missing one legend, < < the best player in his R-squared 0.799 0.799 tor y, > > Maradona . It’s number 664. Controls and Vendor FE Yes Yes < < The guy is a genius! > > How much is it? Errors Standard Robust Okay. Let’s continue from the beginning. I need two Number of observations 1712 1712 from the front page. *p < .1. Standard errors are shown in parentheses. The dependent variable in all Number 2, which is the Trophy; and number 7, which specifications is the vendor’s price quoted in thousands of Colombian pesos is the Ball. How much for each one? I also need one sta (kCOP). Controls in all regressions: geographic cluster, day and sticker fixed effects, (perceived) age and gender of the vendor, whether the vendor has dium, number 11/12. How much is it? nearby vendors, and journey I need (the picture of ) one team, it is Denmark/Ice- land, number 241/281. How much is it? And finally, I need one player, number 129/574. It’s a the printed list. Hide the printed list and do not get too player from Spain/England. How much is it? close to the vendor.] [Annotate all the prices in the hand-written list and Good morning/afternoon, how are you? terminate the interaction.] My husband/wife and I are looking for the missing Thank you! I may come back, or my husband/wife will stickers to complete our kid’s PANINI album. come back. At what time of the day can I find you here? He/She is also looking for the stickers right now, so Thanks again, have a good day. for the moment, I am only quoting. Do you complete [Walk away from the seller and write all the prices on lists? the printed list.] [If the vendor says “Yes,” continue the script. Other- wise, say goodbye and look for the next vendor ID.] Script #2: Short list Since the protocol is very similar, we OK, I tell you which ones I need, and you tell me the only provide below the part of the script concerning the price. quotation of the four stickers in the Short List treatment. I am going to write them down because they are for Unlike the Long List treatment, in this case we quote stick my kid, and I also have to talk to my husband/wife. ers in ascending order according to the album’s numbers. [Start writing down the prices on the list assigned We altered the order because it would be easier for the to that vendor. If the vendor does not have a sticker, Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 19 of 24 6 Fig. 2 Sampling of sticker vendors in five clusters in Bogotá. The maximum distance between clusters is 15 km. Panels on the right side represent an 800 × 800 m grid for each cluster 6 Page 20 of 24 P. Zamora et al. 59 Vendors, 287 Obs. C-24 C-23 C-22 C-21 C-19 C-18 C-16 C-15 C-14 C-13 C-12 C-11 C-10 C-9 C-8 C-7 C-6 C-5 C-4 C-3 C-1 S-14 S-13 S-12 S-11 S-10 S-9 S-8 S-7 S-6 S-4 S-3 S-2 S-1 T-9 T-8 T-7 T-6 T-5 T-4 T-3 T-2 T-1 P-11 P-9 P-8 P-7 P-6 P-5 P-3 P-1 K-13 K-11 K-10 K-7 K-3 K-2 K-1 N-1 1. Arg-F-LL 2. Arg-M-LL 3.Col-F-LL 4. Col-M-LL 5. Col-F-SL 6. Col-M-SL LL (199 obs.) SL (88 obs.) Fig. 3 Mapping between buyers and vendors. Columns represent each one of the buyers, and rows represent each one of the sellers. Circles denote a vendor–buyer interaction in the Long List treatment, and hollow diamonds denote a vendor–buyer interaction in the Short List treatment. vendor to look for the stickers, and in the Short List treat- I need the Panini Knight Logo/FIFA Fair Play, it’s num- ment, there was no need to prime the buyer’s nationality. ber 0/1. How much is it? I also need the Trophy, number Okay, let’s start with the emblems on the first page. 2. How much is it? Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 21 of 24 6 Stadium Team Player 100 100 100 80 80 80 60 60 60 40 40 40 20 20 20 0 0 0 0 1 2 3 0 1 2 3 0 1 2 3 kCOP kCOP kCOP Colombians Argentinians Fig. 4 Cumulative distribution of quoted prices. Panels from left to right represent the cumulative distributions for the stickers with stadiums, teams, and players. The distribution for Colombians is displayed in blue dots. The distribution for Argentinians is displayed in red triangles. Good. I also need the logo of Brazil/Germany, it’s num- • Do you share your business with someone else? Fam- ber 340/420. How much is it? Finally, I need the logo of ily? Partner? Friend? Employee? Colombia. How much is it? • At what time do you sell more? • How did you establish your prices? • Do you get upset if people tell you that they are only Appendix 3 going to quote? See Table 10. • Do you change the prices throughout the day? • Do you charge more per sticker to some people than Questions for the structured interview others? Women with respect to men? Foreigners with Questions in parenthesis were only asked when applied. respect to locals? To young people or to older peo- ple? • What is your usual work schedule? Why that sched- • Any difference in prices before it started and during ule? the World Cup? • Does anyone else work here with you? Are you usu- ally accompanied? Table 10 Schedule of conducted interviews Date Number of Interactions Time Zone Gender Perceived age 6 1:30 p.m Street 93rd Female Old woman July 2, 2018 (holiday) 6 2:00 p.m Street 93rd Male Young man 6 3:00 p.m Street 147th Female Middle Age 6 3:30 p.m Street 147th Male Middle Age 6 9:50 a.m Downtown Male Middle Age 5 10:15 a.m Downtown Female Old woman July 3, 2018 (business day) 6 11:10 a.m Street 72nd Female Middle Age 6 11:40 a.m Street 72nd Male Old man 6 12:00 p.m Street 85th Male Middle Age 6 12:10 p.m Street 85th Male Middle Age July 13, 2018 (business 4:30 p.m Street 147th Female Middle Age day) Percent 6 Page 22 of 24 P. Zamora et al. • When were the highest prices reached and when el dicho... al mejor postor, ¿no?... entonces pues diga- were the lowest? mos, a los que se les cobra un poquito más o algo así, • Do you sell or used to sell counterfeit stickers? de pronto sí son los extranjeros.” • How can you find the counterfeits? Who sells them? [Excerpt 5] “En el mundial pasado había varios • Why are you located here? equipos que salían mucho y otros que no salían. En • (Why are you so close to the supermarket?) cambio este año salieron todos los equipos, todos los • (If you know there are many vendors here, why not jugadores. Lo que no salieron fue las doradas. Diga- go further?) mos, de una sola caja, estaban saliendo por ahí 18, • For how long have you established your business 19 escudos. Son 50 en total, entre primera, última here? página y escudos. Salían 18, por mucho salían 20, • What do you do outside the World Cup season? Do 22, ¿si ves?” you sell something else? [Excerpt 6] “Pasan preguntando: “necesitan escu- • Does the police often come to carry out inspections? dos?”, yo contesto: “sí, pero originales”, entonces ellos What do you do when the police arrive? dicen: “Nooo, usted sabe que yo no tengo de esos”, • Do you pay, or paid before, social security? Pension entonces respondo: “¡ahm no!, yo ya tengo de esos, contributions? Health insurance? necesito originales”. Eso venden el paquete de 50 • Have you ever rented a place to sell your merchan- láminas a 25 o a 20” dise? (How much do you pay for rent?) [Excerpt 7] “A mí por el local me toca vender sólo lo • Why did you decide to dedicate yourself to selling que es original, pero en la calle hay mucha lámina things in this way? escaneada que los vendedores están ofreciendo y hay • Do you live far away? gente que no conoce. Entonces, ¿qué pasa? a mí un • What do you do with the merchandise after the jour- día me pasó un inconveniente con un funcionario ney? Where do you leave it? de la DIAN. Era un buen cliente y é me traía com- pañeros. Resulta que yo por equivocación y por estar allá (señala la calle) y estar en la librería, recibí Original Excerpts in Spanish unas láminas y yo las intercambié y después las iba a vender. ¡Resulta que precisamente se las di a é! y el [Excerpt 1]“Pues es que ya uno tiene un recorrido funcionario ¡me trajo la policía!... pero yo no sabia, con el mundial, ¿sí?... que por lo menos uno coge y yo no sabía que las láminas eran falsas. Es que son destapa cajas… porque es que uno destapa... tiene muy bien hechas, pero sin embargo tiene un detalle es que destapar cajas o pacas como se dice... porque que uno puede reconocer.” una paca trae 12 cajas. Entonces pues digamos uno [Excerpt 8] “Mire yo soy muy honesta con el cliente. destapa 3, 4 pacas. Entonces, ya arreglando las Yo acá no trabaja las chiviadas, pero a mí me pedían fichas por números y todo, ahí uno deduce cuál es chiviadas... y ahora yo traigo chiviadas. Entonces yo el equipo que más sale, los jugadores que más salen, le digo al cliente: < < bueno, quiere barato o quiere las fichas que menos salen. Entonces de ahí los que caro?, ¿quiere original o no original? > > ... entonces más saca son los más baratos y los que menos salen ya depende del cliente. Le digo: < < bueno, ya a con- son los má costosos.” ciencia lleva chiviada... > > ” [Excerpt 2] “Establecimos los precios de acuerdo a la [Excerpt 9] “Es dependiendo... por lo menos ahorita experiencia de ventas anteriores y averiguando con en el mundial, a medida que va ganando Colombia, otros vendedores.” va subiendo.” [Excerpt 3] “Pues yo he tratado de cambiar los pre- [Excerpt 10] “Hace cuatro años las camisetas estu- cios, porque por lo menos con las láminas los escu- vieron escasas. Ganó Colombia y se subió el precio... dos son los dif íciles y pues ahí yo soy la que má caro enseguida subió el precio. Entonces yo dije, este año vende. El resto sí por la guerra de precios, entonces va a pasar lo mismo, y mentiras. A mí me hicieron se venden barato. Es el que más vende… no es el un préstamo de 10 millones de pesos y lo invertí que más gane, sino el que más vende... el que quiera en un poco de camisetas. La pagué a 23,000 pesos pelear a un cliente, el que le dé más barato, aparte y hoy fui a comprar unas chaqueticas para niño y que los clientes le ponen precio a la mercancía... o me doy cuenta que está a 4,000 pesos... ¡ya qué!... ya sea uno ya no puede ponerle precio, porque ellos puedo llorar sobre la leche derramada. Si ganan voy vienen ya a ponerle el precio a uno.” y compro. Ahí repongo algo de lo que... terrible... ter- [Excerpt 4] “No, de por sí, no... el precio es como el rible... ¿sabe qué es eso? mire, compramos camiseta mismo. Lo que cambia es como se dice... como dice a 23,000 pesos y la estamos vendiendo a 25,000... ¿se Price discrimination in informal labor markets in Bogotá: an audit experiment during the 2018… Page 23 of 24 6 Ayres, I., Siegelman, P.: Race and gender discrimination in bargaining for imagina sumercé esperar cuatro años para ganarse anew car. Am Econ Rev 85, 304–321 (1995) uno en una camiseta 2,000 pesos?” Azmat, G., Petrongolo, B.: Gender and the labor market: What have we [Excerpt 11] “Mi puesto era en la puerta del Banco learned from field and lab experiments? Labour Econ 30, 32–40 (2014) Baert, S., De Pauw, A.-S.: Is ethnic discrimination due to distaste or statistics? Bogotá y me enferme´ y me fui para la casa. Ahí en Econ Lett 125(2), 270–273 (2014) la bahía había una cosa de las zonas azules y ahí Balafoutas, L., Kerschbamer, R., Sutter, M.: Second-degree moral hazard in a yo tenía un puesto, después como me dio cáncer real-world credence goods market. Econ J. 127(599), 1–18 (2015) Balafoutas, L., Nikiforakis, N., Rockenbach, B.: Direct and indirect punishment me fui pa’ la casa. Ahora no vengo sino a traba- among strangers in the field. Proc Natl Acad Sci. 111(45), 15924–15927 jar los mundiales acá. [Entrevistadora: ¿y es mejor (2014) estar así rodeado? Porque aquí generalmente hay Baldini, M., Federici, M.: Ethnic discrimination in the Italian rental housing market. J Hous Econ. 20(1), 1–14 (2011) más gente, ¿no?] ¡Muchísima! Mire sumercé, yo Balsa, A.I., McGuire, T.G.: Statistical discrimination in health care. J Health le digo una cosa: lo que Dios le tiene a uno, se lo Econ 20(6), 881–907 (2001) guarda, así esté uno donde esté.” Baranzini, A., Schaerer, C., Ramirez, J.V., Thalmann, P.: Do foreigners pay higher rents for the same quality of housing in Geneva and Zurich? [Excerpt 12] “Sí, nos dejan trabajar aquí con la Swiss J Econ Statist 144(4), 703–730 (2008) condición de que no dejemos meter a ninguno Becker, G.S.: The economics of discrimination. University of Chicago Press, nuevo.‘” Chicago (1957) Bertrand, M., Chugh, D., Mullainathan, S.: Implicit discrimination. Am Econ Rev 95(2), 94–98 (2005) Acknowledgements Bertrand, M., Mullainathan, S.: Are Emily and Greg more employable than We gratefully acknowledge support from Santiago Sautua, Yuliet Verbel, and Lakisha and Jamal? A field experiment on labor market discrimination. Juan Baquero during the execution of the field experiment. We also thank Am Econ Rev 94(4), 991–1013 (2004) Diego Aycinena, Francesco Bogliacino, Simeon Schachtele, and participants in Castillo, M., Petrie, R., Torero, M., Vesterlund, L.: Gender differences in bar - the LACEA-BRAIN workshop for their valuable comments. gaining outcomes: a field experiment on discrimination. J Public Econ The information and opinions presented herein are entirely those of the 99, 35–48 (2013) authors. Financial Support from the program “Inclusión productiva y social: Charness, Gary, and Peter Kuhn. 2011. "Lab Labor: What Can Labor Econo- programas y políticas para la promoción de una economía formal, código mists Learn from the Lab?" In Handbook of Labor Economics , Vol. 4, 60185, que conforma la Alianza EFI, bajo el Contrato de Recuperación Contin- ed. Orley Ashenfelter and David Card, 229-330. Amsterdam: Elsevier. gente No. FP44842-220-2018.” is gratefully acknowledged North-Holland. Chen, L.-Z., Wei-Min, Hu., Szulga, R., Zhou, X.: Demographics, gender and Authors’ contributions local knowledge—price discrimination in China’s car market. Econ Lett All authors have participated in (a) conception, design, and analysis of the 163, 172–174 (2018) data; (b) drafting the article or revising it critically for important intellectual Csizma, K.A., Wittig, A.F., Schurr, K.T.: Sport stereotypes and gender. J Sport content; and (c) approval of the final version. All authors read and approved Exerc Psychol 10(1), 62–74 (1988) the final manuscript. Currie, J., Lin, W., Zhang, W.: Patient knowledge and antibiotic abuse: evidence from an audit study in China. J Health Econ 30(5), 933–949 Funding (2011) Financial Support from the program “Inclusión productiva y social: programas y Fang, Hanming, and Andrea Moro. 2011. “Theories of Statistical Discrimination políticas para la promoción de una economía formal, código 60185, que conforma and Affirmative Action: A Survey.” Pp. 133–200 in Handbook of Social la Alianza EFI, bajo el Contrato de Recuperación Contingente No. FP44842-220- Economics, Vol. 1A, edited by J. Benhabib, A. Bisin, and M. O. Jackson. 2018.” is gratefully acknowledged. Amsterdam: North-Holland, Elsevier. Flage, A.: Ethnic and gender discrimination in the rental housing market: Evi- Availability of data and materials dence from a meta-analysis of correspondence tests, 2006–2017. J. Hous. Data and code are available in an Open Science Foundation repository: https Econ. 41, 251–273 (2018) ://osf.io/gzhfs /. GrabowKu¨ hl, H.M.: You don’t bend it like beckham if you’re female and reminded of it: stereotype threat among female football players. Front Competing interests Psychol 10, 1963 (2019) The authors have no affiliation with any organization with a direct or indirect Grossman, S., Honig, D.: Evidence from Lagos on discrimination across ethnic financial interest in the subject matter discussed in the manuscript. and class identities in informal trade. World Dev. 96, 520–528 (2017) Grytten, J., Skau, I., Sørensen, R.: Do expert patients get better treatment than Author details others? Agency discrimination and statistical discrimination in obstetrics. Department of Economics, Texas A&M University, College Station, TX, USA. J Health Econ 30(1), 163–180 (2011) Department of Economics, Universidad del Rosario, Bogotá, Colombia. Kim, A.M.: The mixed-use sidewalk: Vending and property rights in public space. J Am Plan Assoc 78(3), 225–238 (2012) Received: 30 October 2019 Accepted: 12 January 2021 Kim, S.-H., Fernanda, L., de Leon, L.: In-group and out-group biases in the marketplace: a field experiment during the World Cup. Oxford Economic Papers 71(3), 528–547 (2018) La Rafael , P., Shleifer , A.: Informality and development. J Econ Perspect 28(3), 109–126 (2014) References Levy, S.: Good intentions, bad outcomes: social policy, informality, and eco- Apollo, M.: Dual pricing-two points of view (Citizen and Non-citizen) case of nomic growth in Mexico. Brookings Institution Press, Washington (2010) entrance fees in tourist facilities in Nepal. Procedia Soc Behav Sci 120, List, J.A.: The nature and extent of discrimination in the marketplace: Evidence 414–422 (2014) from the field. Quart J Econ. 119(1), 49–89 (2004) Arrow, Kenneth. 1973. The theory of discrimination in labor markets. In: O. List, J.A., Rasul, I.: Field experiments in labor economics. In: Handbook of labor Ashenfelter, A. Rees (Eds.). Princeton, NJ: Princeton University Press. economics, vol. 4, pp. 103–228. Elsevier, Amsterdam (2011) Auspurg, K., Hinz, T., Schmid, L.: Contexts and conditions of ethnic discrimi- Martínez, L., Short, J.R., Estrada, D.: The urban informal economy: Street ven- nation: Evidence from a field experiment in a German housing market. dors in Cali, Colombia. Cities 66, 34–43 (2017) J Hous Econ. 35, 26–36 (2017) 6 Page 24 of 24 P. Zamora et al. Mullainathan, S., Noeth, M., Schoar, A.: The market for financial advice: an audit Webster, C.: Property rights, public space and urban design. Town Plan Rev 78, study. National Bureau of Economic Research, Cambridge (2012) 81–101 (2007) Oxford, S., McLachlan, F.: “You Have to Play Like a Man, But Still be a Woman”: Yip, C.M., Wong, R.-K.: Gender-oriented statistical discrimination theory: Empiri- young female colombians negotiating gender through participation in cal evidence from the Hong Kong labor market. Res Soc Stratif Mob 37, a sport for development and peace (SDP) organization. Sociol Sport J 43–59 (2014) 35(3), 258–267 (2018) Siegelman, P., Heckman, J.: The urban institute audit studies: their methods Publisher’s Note and findings. Clear Convinc Evid 187, 258 (1993) Springer Nature remains neutral with regard to jurisdictional claims in pub- Turchin, Peter. 2016. Ultrasociety: How 10,000 Years of War Made Humans the lished maps and institutional affiliations. Greatest Cooperators on Earth. Chaplin, CT: Beresta Books.

Journal

Journal for Labour Market ResearchSpringer Journals

Published: Mar 14, 2021

There are no references for this article.