Skip to main content

Relationships as a market

In this blog, I will examine relationships abstractly viewed as a market. Specifically, it will focus on intimate relationships that will be taken to span attraction, dating, courting and marriage phases between couples. The agents are people and a couple should be restricted to two agents who are compatible with each other. In this market, the currencies for exchange may range from emotion (love), security, validation, reputation to satisfaction among others. Among these currencies, we will focus on security and love as the item being exchanged in the market assuming they come as a package. The assumption is based on my individual bias that the two items can never be mutually exclusive for a couple to function well. Also, it’s based on interpreting security as a provision that one party in a couple provides that the other party needs but cannot individually “generate” to satisfy them and vice versa. This choice of market is based on my interest in psychology of intimate relationships. Mainly why intimate relationships should be viewed as partnerships instead of solely love or security driven “adventures” as they usually are.


This market could be modeled based as a love-security combination as the main design feature. Assuming that each member of any particular couple has a preference list of all possible partners they are compatible with based on the love-security measure, we create a networked market model whose goal is to match parties to maximize satisfaction per the measure. We could create a mechanism based on assumptions including but not limited to assuming parties are initially single, dating/married (that is have an initial endowment), have weak or strict preferences that may be complete or incomplete or no preferences at all (for people who prefer being alone) to find matchings in this market. If we were to introduce money in this market as the only currency, then it leads to relationship dynamics that are existent in the current society.


Alternatively, we could use reputation or other societal-measures as the main feature to measure satisfaction and design mechanisms to create matchings. This would be assuming that whatever society gives approval to would be the basis with which the agents will determine their preference list. In general, there are many other features that could be used to design this market as might be reflected on dating sites but this model is meant to extrapolate the idea to a personal level even to those without access to technology.




The market for online video content – YouTube

The market that I have chosen to study is the market for online media content on the video streaming platform YouTube. In this market, the “sellers” are the content creators on the platform. In order for these people to be successful they must look to differentiate themselves in the market for content by developing their own niches in terms of the type of content they produce. They also must develop a level of trust with their subscribers, and commit to posting content on a consistent basis. The “buyers” in this market are the people visiting the YouTube website and viewing the content. A given buyer’s favorite content creators and genres of videos can be thought of that buyer’s preference profile, where their favorite types of videos are preferred over their least favorite types of videos. Finally, the exchange that occurs is content and videos produced by the content creators are exchanged for views from the population of viewers watching videos on YouTube. Because of this, the currency in this market can be thought of as views. While views on a video is by no means a true form of currency, it does translate into actual monetary profits dependent on how many views a content creator may get. According to USA Today1, YouTuber’s earn a portion of the advertising revenue that their videos generate for YouTube, and the top 12 content creators on YouTube made a combined $70.5 million in 2016 alone. Therefore, more viewers consuming your content leads to more ad revenue being generated for yourself. This new style of market where anyone can enter and be successful as long as they’re creating content that enough people enjoy has led people to make a viable living by just producing content that they love. These are people that without YouTube would likely be having a career that makes them feel much less excited and creatively stimulated, which is why I am so interested and passionate about this network economy.

An interesting aspect about the design of this market is the decision to have an open platform where anyone can join and be a creator. When you think about other forms of video content that a person may consume (i.e. television or movies), it is very difficult and competitive to be involved in the creation of this content. Actors and actresses must audition, and writers and directors must make their case for a studio head in order for their project to be funded. On the contrary, any person with a computer can create a YouTube account and upload a video. From an efficiency standpoint, this design choice is a massive advantage for YouTube, since it has to put in no effort in order to locate talent. The public will select the most talented content creators by viewing their videos the most, and YouTube’s algorithm will simply funnel that creator a certain portion of the advertising revenue. One potential setback of this design choice may be that the platform gets flooded with poor-quality content since anyone can post anything. However, YouTube has mitigated this by featuring the most popular videos on the front page, and removing content flagged as inappropriate in any way. The design choice to allow anyone to become a content creator allows YouTube to skip the talent-sourcing process and organically add content creators that are passionate about producing the best content possible.

In a world where more and more people are switching from watching traditional cable TV to streaming video online, YouTube has positioned itself perfectly to facilitate the market for online media content by organically sourcing content creators to produce content for their massive viewership to enjoy.



Baseball card market

A market many kids participate in is the ‘baseball card market’. This is the market in which children (agents) have baseball cards and are willing to trade said cards with each other. In the market I’m defining there is no real money, kids only trade cards for other cards. Because of this the currency is the cards themselves.


The market design in this market has several interesting properties. Each agent can own many baseball cards. In previous examples we have seen in class each agent only gets one recourse but in this each agent can have several. Another aspect of this market is that trades in the market don’t have to be one to one. For example, someone can trade 5 cards for one particularly valuable card. This market of course allows each agent to have preferences over all possible cards. When one considers that trades aren’t one to one its important to not only give agent preferences over all cards but to have agent preferences over all sets of cards.


Because preferences are over sets of cards it creates a really convoluted market. There would still be parento-efficient matchings but every other aspect of the market is significantly more complex then what we have seen so far. Serial dictatorships won’t work because of course everyone’s first preference is to have all the cards. In fact I can’t really think of a mechanism that fairly allocates cards and is strategy proof.

College Admissions as a Market: Tournaments, Revealed Preferences, Affirmative Action

Every year, millions of (typically) high-school seniors submit tailored college applications, entering competition pools at higher-education institutions of their choosing. This well-known market of college admissions is two-sided – college admissions officers, reactively, extend invitations to candidates they deem as having the potential to contribute to their university’s continued success (via perceived ability/desirability and willingness/likelihood of attendance given acceptance) and accepted applicants decide which college (among invitations) they prefer the most.

In this scenario, the colleges a student applies to constitute the students initial preferences: each of these schools as a non-zero utility if the matching is realized. These non-zero utilities act as the good/service given by colleges – it is the value that they can offer matriculating students. Acceptances form the set of edges that are possible matches, and students typically choose the school which has the highest value. The currency by which the matching/service is realized is acceptance of the offer letter (assume tuition is a consideration in the decision to accept). These exchanges can only be meaningfully made between applicants and colleges.

Market design – The rule used to allocate acceptance offers is the college’s criteria for admissions. GPA, extracurricular activities, personal statements, recommendations, financial capacity, connections, how much a student wants to attend the school (e.g. Early Decision), etc. are all taken into consideration. Furthermore, a student’s race is also taken into consideration for affirmative action. Alternatives simply include different admission criteria.

Question to explore: Do rankings beget preferences or do preferences beget rankings?

Implications extend most readily to the job market – e.g. affirmative action for veterans, minorities, and those with disabilities.

Note: I wanted to explore the thought of considering one person as the amalgam of multiple economic agents i.e. the planner versus the doer, the investor versus the indulger, the sophisticated versus the naïve consumer. One example of a scenario in this market would be decisions to smoke (given that s/he values life, the planner would want to avoid smoking while the doer would prioritize immediate gratification). The market would consider the difference in utility (discounted over time – e.g. with unnoticed present bias for naïve consumers) individuals receive between each option, preferences, etc. However, I thought this might be a reach – just want some feedback for future consideration!

Sources: – Christopher N. Avery & Mark E. Glickman & Caroline M. Hoxby & Andrew Metrick, 2013. “A Revealed Preference Ranking of U.S. Colleges and Universities,” The Quarterly Journal of Economics, Oxford University Press, vol. 128(1), pages 425-467.

The MLB Draft

The MLB Draft lasts for 40 rounds and the teams are allowed 1 pick per round.  The teams are given picks in the order of worst record to best record from the previous season’s standings. This allows the worst team to get the first pick so they have the best chance to improve in the future, and the best team to get the 32nd pick in every round.  The asset (people shouldn’t be considered currency) being exchanged is rights to sign future MLB players. This asset is technically the right to sign a player, so after being drafted, the player can negotiate their contract and decide if they want to sign or return to college.  Teams are able to trade with each other, both for picks in the current draft, future drafts, and current players they have on their roster.

Unlike in the other major American sports leagues, players in the MLB draft who have not gone to college yet, after being drafted by a team, can elect to go to college for 3 years, and then reenter the draft and get drafted by a different team.  This is an interesting design choice, because it gives the players significantly more power than in other sports.  Since a player doesn’t want to play for the team they are drafted by, they can go to college and come back in 2-3 years to play for a different team, because the original team cannot draft them unless they consent to it.  This affects how teams draft their players because they will be much less willing to use one of their highest picks on a player unless they know for sure that the player in question will sign with them.  A change to this market would be the players would get drafted and then be able to go to college but teams would still retain the rights to the player like in the NHL.  This would give teams more power in the market because players drafted by teams they don’t want to play for would still have to play for those teams.

Modern Romance: Online Dating as a two-sided market

The dynamics of dating have been discussed and dissected extensively, but in this post we can hopefully gain some new insights into it by viewing it as a networked market. Online dating specifically, renders itself quite naturally as a market (albeit a non-traditional and subtle one) and as of 2015, up to 38% of single Americans report having participated in it[1]. At its most abstract, online dating is a network connecting people participating in a two sided market where potential partners can “match” with each other. What makes this market quite interesting is that there is no explicit currency, price, or transfer mechanism. The transaction in question is a social transaction, not an economic one, and can be viewed as an exchange of social/ psychological rewards or an allocation of relationships. If we think of currency as social currency based on attractiveness or desirability, we get an interesting market in which even though currency is required for transactions and more currency can lead to more desirable outcomes, the currency is never actually “spent” as the exchanged “good” itself is indivisible.


This idea of currency itself is dynamic as it is a function of both the individual’s attributes and the potential partners’ preferences. Hence, an agent may have high currency or desirability for one person and low desirability for another, and the preferences may not necessarily be monotonically related to their attributes. This makes matching in the market quite interesting as individual preferences are likely to be heterogenous. Efficient matching in this market thus relies on the existence of pairs of mutually desirable agents in a setting where all preferences are heterogenously distributed. An interesting question one may ask is how, then, is the efficiency of dating markets so high in real life? One possible explanation, as Hitsch, Hortacsu and Ariely suggest[2], might be that there is natural sorting in dating markets based on attributes, and also possibly because people tend to prefer partners who are “similar” to them, and thus the market naturally resolves into pairs of mutual desirability. An interesting sidenote is that if our previous assumption of heterogenous preferences breaks down, it would indicate that preferences are in fact monotonically related to attributes, and certain attributes are more “universally preferred” than others.


In creating this market, the network effect is fundamental. The main appeal of online dating is the availability of many potential mates, and achieving a critical mass is key to the success of an online dating platform. Agents’ utility increases if there are more users on the “other side” of the market, as there is a higher number of potential transactional partners, and this is a positive cross-side effect[3]. However, as more users join “their side”, users’ utility decreases due to higher competition, and this is a same-side negative effect. In general, additional users add to congestion and increase search costs (along with competition if on the same side). Thus, the efficiency of the network depends not just on number of users, but also on their potential to be a match for others, and a good strategy for the platform might be to limit users to those likely to find matches.


Because of the importance of network effects to the efficiency of the market, it is common to see design choices made to exploit this factor. By their very nature, creating online platforms involves making embedded design choices which influence not just potential matchings, but also structure information in a way that deliberately creates asymmetry. With regards to network effects, a common design choice on online dating platforms is to conceal knowledge about participants, so the total number of potential partners is hidden and only a limited number of recommended partners are displayed everyday. This artificially boosts the cross side positive effect, as it creates the illusion of unlimited potential partners somewhere in the network (as a sidenote, this technique is also utilized quite effectively by Netflix, and you’ll never see the total count of movies on their website). Similarly, the users are also not told how many other users are on “their side”, and withholding this information hides the true severity of competition in the market.


The platform operator may also choose to subsidize participants on one side of the market, as was the case with Ashley Madison, which offered free subscriptions for women[]. This helps balance the split of the market, keeping both sides interested. Another debatable design choice is the use of recommendation algorithms vs letting users search for partners. While searching encourages users to filter through all users and attempt to form matchings according to their own beliefs over their preferences, recommendation algorithms take the opposite approach and try to present a limited set of potential matchings based on inferred user preferences. The search vs recommend design decision also determines knowledge other users have of preferences – while the search design allows users to directly observe each others’ preferences, the recommendation design forces users to only infer preferences. We can argue that recommendation algorithms are more efficient, as they would only show users to each other if they believed that both users could find each attractive, and thus they would reduce search costs. Moreover, limiting the view of the market to just a few options at a time may also make those options seem more attractive. However, this leads to certain agents never even knowing about the existence of others, and heavily impacts the final matchings agents end up with.


There is also a considerable amount of information asymmetry on both sides of the market, as users have an incentive to present a biased view of themselves on their online profiles. Furthermore, design decisions may actually encourage information asymmetry, such as in the case of Tinder, on which matches are judged based on a few pictures and minimal profile information. Since the app is also designed to be “fast-paced” and displays many potential matches in quick succession, it encourages users to find a breadth of potential matchings rather than depth.


Finally, we must ask ourselves the question, is the matching produced by online dating services actually “good”? If we define “good” to be each agent getting their top choice, then it is unclear, as we can only observe a binary view of preferences rather than rank-order preferences. However, if we define it to be a “stable” matching, then evidence suggests that these are actually pretty good matchings. In their 2010 paper[4], Hitsch, Hortaçsu, and Ariely use the Gale-Shapley algorithm based on estimated male preferences to predict the sorting patterns found in online dating. They find a significantly strong correlation, and conclude that the stable matching predicted by the Gale-Shapley algorithm can be seen as the limit outcome of this two-sided search and matching model. Due to the properties of the Gale-Shapley algorithm, this has some powerful implications, such as that the matching produced by online dating is a stable matching, and is also Pareto-optimal match, within the set of stable matches, for the side of the market (men or women) that makes the offers in the deferred acceptance procedure.


The correlation of online dating outcomes with the predictions from stable matching algorithms definitely indicate their efficiency and usefulness, but we have also seen several sources of inefficiency and information asymmetry in the market. Perhaps a reasonable improvement could be a decrease in information gaps, maybe by making preferences more transparent and profiles more informative. Another improvement could be finding a better balance between searching and recommendations, so that users have more control over matchings but are still shown only what is most relevant to them, in order to reduce search costs. Although flawed and overly reliant on design and implementation choices, online dating is still a complex and multi-faceted network that warrants further discussion.


– Bhai Jaiveer Singh



1) Ansari, Aziz, & Klinenberg, Eric. “Modern Romance.” New York: Penguin Press, 2015.

2) Burtch,Gordon & Ramaprasad, Jui. “Assessing and Quantifying Local Network Effects in an Online Dating Market.” Working Papers, NET Institute, 2016.


4) Hitsch, Günter J., Hortaçsu, Ali, & Ariely, Dan.”Matching and Sorting in Online Dating.” American Economic Review, 2010.


[1] Ansari, Aziz, & Klinenberg, Eric. “Modern Romance.” New York: Penguin Press, 2015.

[2] Hitsch, Günter J., Hortaçsu, Ali, & Ariely, Dan.”Matching and Sorting in Online Dating.” American Economic Review, 2010.

[3] Burtch,Gordon & Ramaprasad, Jui. “Assessing and Quantifying Local Network Effects in an Online Dating Market.” Working Papers, NET Institute, 2016.


[4] Hitsch, Günter J., Hortaçsu, Ali, & Ariely, Dan.”Matching and Sorting in Online Dating.” American Economic Review, 2010.

Market for Office Furniture

Context: All offices purchase furniture, generally this is done by having office staff submit requests to their organization (a university department, or other governing body which manages other offices as well). In many cases, these organizations allocate money to office needs based on historical costs.

I think it is useful to analyze the market for office furniture at two different scopes: intra-organizational and inter-organizational. First I will analyse the intra-organizational market:

The allocating body has some pool of resources allocated towards furniture and absorbs all costs; no currency is used with respect to offices.

Every office has a value function, and therefore a preference ordering, over furniture. Given this preference ordering, they request furniture from the allocating body. I am not sure what sort of allocation schemes organizations generally use, I can assume that it is some function of the office’s reported preferences and the item’s cost. I assume that any allocation process that the organization cannot fulfill (for lack of resources) will be reset, allowing offices to re-represent preferences. However, I can show that no allocation scheme in this setting is strategyproof. An office can always report exaggerated value for its most preferred items, and exaggerated lack of value / dispreference towards less preferred items.

  • If offices are asked to list satisfying items, offices who list more expensive items will be given more expensive items than those who list cheaper ones.
  • If offices are given a certain amount of currency specific to the market (furniture dollars), they will spend it all and never report an ability to have spent less
  • If offices are asked to evaluate items, offices can report manipulated preferences freely, so any scheme designed to maximize (value – cost) will fail to be strategyproof.

I would argue that the inter-organizational network is not a meaningfully networked economy. This is because:

  • Office furniture is priced via quote (that is, prices are not publicly visible or universal)
  • Office furniture is usually very expensive to transport, so organizations generally do not trade among themselves
  • Organizations are not equipped to bid on furniture in a cost-minimizing way, further reducing an incentive to trade among themselves or prefer transparently priced items

College Recruitment as a Market

The entire process of college recruitment is extremely complex with hundreds of variables. Ultimately it can be generalized as a market where students allocate their tuition and presence to institutions that allocate their education and reputation. Universities give out acceptance letters based on students who meet certain criteria and are deemed acceptable. On the other side, students pick which university they would like to attend through preferences that are (usually) non-strict and characterized by a categorical ranking – ‘Safety Schools’ <= ‘Target Schools’ <= ‘Reach Schools’.

There is an exchange of monetary currency only from the students towards the universities however there is also the currency of reputation that the university offers in the form of a diploma as well as quality of education, both of which are large factors for students deciding where they want to go. The topic of ‘currency’ in this market is very interesting with both sides holding different forms and each party valuing that currency different based on their needs and wants. For example, take a star basketball point guard that gets a full athletic scholarship from UNC and Harvard University, and must now choose which school to attend. Harvard has arguably the better alumni connection, reputation, education, and endowment, yet the student could very likely choose UNC as they were the 2017 NCAA basketball champions and he sees himself getting the best chance of being drafted into the NBA at UNC. In the end, where a student decides to go may not be as easily quantifiable as simple college rankings but could rather be dependent on their intrinsic values and goals.

A design feature in this market is the acceptance rate which is often times manipulated to some degree by universities. In this market, apart from gaining tuition as a currency, universities also see students as a currency, with students that excel in their respective programs being of higher value. Stronger students lead to higher chances of success rate upon graduation which usually benefits the university either through donations or simple improvements in reputation. Strong students are often times influenced by university rankings which is in turn is influenced by acceptance rates. Thus, you see top tier universities like Harvard, Stanford, and Yale battling over who has the lowest acceptance rate in order to claim that top accolade of being the most selective university in the world. This has come to a point where normal schools have caught on and have purposefully rejected students that they believe are ‘too good’ to attend in order to lower their acceptance rates.

Beyond the Court: The Trade Market in the NBA

I love to watch sports; it is one of my activities to do when I am not worrying about final projects, prelims, and extracurriculars at Cornell. Sports are generally seen as a form of escapism from our current realities, but what many don’t see on the football fields and the basketball courts is the very intrinsic but complex trading market that manages the quality of play of many of our favorite sports teams. In professional basketball, specifically in the National Basketball Association (NBA), trading is a big part in determining how good your team will be in the short term, and/or in the long term.

A market like this works like a basic exchange market, in which assets are exchanged between agents based on multiple factors such as weak preferences and initial endowments. The agents are in this case are the teams, specifically the General Managers (GMs), but the ‘assets’ being traded are what make the NBA market (and other sports market) so unique. There are multiple kinds of assets that can be exchanged between teams with each type having unique to each team. In the NBA, assets can be players, draft picks (to draft other players), incentives, coaches (in rare cases*), and, of course, money.

Hw the market works is like: say there is a team with a player (or players) that is not a good fit for the team and needs to be traded before the deadline. What happens next is that the team starts to contact possible trade partners who are willing to take part in an exchange (there can be multiple partners so this market is not always binary). When they find a team that is willing to trade with them (or has assets that they want), negotiations take place and the teams discuss what is being exchanged. If they come to an agreement about the terms of the deal, then the deal is finalized and is sent for approval to the league office (NBA). The deal is finalized and the assets are traded (examples **).

In a team, the value of each player can really dependent on a number of things: fit with the team, star value (popularity), relationship with managing, and skill. These factors also determine what players teams go after in a market, and as such preferences for each team tends to be weak; multiple players can have similar values and thus teams are willing to be accepting if they don’t get their top choice. What is mind-boggling, however, is that there are assets that are restricted, or forbidden, from being trade due to their high value on the team. So for instance, star players that bring in a lot of wins and money to a team (Lebron James, Stephen Curry, etc.) will never have their teams even entertain the possibility of a trade, because there is no value in the market that can equate to them (unless it’s another star player, which seems pointless to exchange for unless it’s a matter of fit).

In general, the NBA trade market is a complex system of exchanging players, money, and draft picks in order to help your team ‘succeed’. Whatever success means for each team is up to them, and so you have a market that doesn’t always have fair or equal exchanges. Some trades work out, many don’t. You’ll never know how much value you get from a trade until after you make it, whether that’s a couple of weeks or a couple of years.


Residency Matching Problem

I have always found the residency matching problem very interesting. In this market there are hospitals and medical students, and both sides of the market have strict preferences. There is no currency in this market, rather our goal is to maximize utility. We want to give the hospitals their top choice residents so that the hospital can continue to be successful. At the same time we want to give residents their top choice hospital because forcing them to live somewhere they do not want to may lead to poor work performance since they will likely be unhappy living there. A good matching pairs residents to hospitals in areas they want to work while still giving the hospitals quality medical students.

The algorithm used to solve this problem is Gale-Shapley’s stable matching algorithm. A very nice property of this algorithm is that, for any matching $M$ of hospitals $h_1,…,h_n$ and residents $r_1,…,r_n$ there does not exist and $h_i$,$r_j$ such that they prefer each other to the agent they were assigned in $M$. This property is what it means for the matching to be stable. This is desirable because if the matching $M$ was unstable, that is if there did exist $h_i$,$r_j$ such that they prefer each other to the agent they were assigned in $M$, then hospital $h_i$ might ditch the resident they were assigned in $M$ and take resident $r_j$ who they prefer and also prefers them to the hospital they were assigned in $M$.

A less fortunate property of the algorithm is that one side is guaranteed to get there best feasible preference while the other side is guaranteed to get there worst feasible preference. This is dependent upon  how the algorithm is implemented. The current algorithm favors hospital preferences, which means the residents will be match with there lowest preference that still results in a stable matching.  Obviously we want the hospitals to get the residents they prefer so they can better serve there community, though it seems unfortunate for the residents. There are no initial endowments, however, there is added complexity at points in the market. For example, two medical students might have gotten engaged or married and therefore want to do their residencies near each other.  This requires special attention during allocation in order to place the two residents near one another.

Kleinberg, Jon, and Eva Tardos. Algorithm design. Harlow: Pearson, 2014.

keep looking »