Effects of Transparent Marketplaces
What happens in markets where consumers don’t know their options, and how such markets become transparent.
This article has 3 parts:
Part 1: The US healthcare review as a non-transparent marketplace. Documented negative effects on consumers.
Part 2: Glassdoor, Yelp/Foursquare/GoogleMaps as examples of crowd-sourced market transparency. Documented effects on consumers and sellers.
Part 3: Trustworthiness problem in crowd-sourcing platforms and potential solutions.
Introduction
Information is power. And some of the 50 zettabytes (1 zettabyte = 1 trillion gigabyte) of data that are hovering in various internet clouds have tangible effects on your income, how much you spend on healthcare, and the quality of your day-to-day lifestyle. But sometimes, the information we need as consumers is either intentionally unavailable to us or simply does not yet exist in a consummable format.
With that in mind, I discuss how having access to certain information has empowered people to make better decisions; and how people can collectively fill in existing information gaps to help each other navigate uncertainty.
To frame the discussion, I start with the US healthcare market as an example of a non-transparent marketplace. I then discuss how nontransparent markets can be transformed through crowd-sourced information transparency on examples of labor market and online rating systems. I finish with a discussion of privacy and information trustworthiness as a rising concern with third-party information aggregators.
Definitions and ideas to keep in mind:
To start, let’s summarize some ideas from economics.
- Market — where two or more parties (sellers and buyers) are involved in some exchange.
- Information transparency — as a buyer you often have the option of checking the price first, before committing to a purchase. Such markets in which price and supply information is readily available to the public are called transparent. [transparent market] I expand transparency to include not only pricing, but any information relevant to the buyer.
- Market efficiency — market (information) transparency is a necessary component for a market to be able to self-regulate — market’s ability to move into an efficient configuration, where no products on the market are overvalued or undervalued, in response to available information [market efficiency].
In short,
- You would not buy a pair of shoes for $200, if you knew (transparency) that a pair in the same condition is available for $50 in the store next door. After all, such behavior is rational.
- Your rational behavior as a buyer requires sellers to also be rational and adjust themselves to remain desirable to you as a buyer (market self-regulates).
Part 1: Non-transparent Healthcare Marketplace
To illustrate how information access comes into play in marketplaces, let’s consider the US healthcare market.
The US healthcare market lacks transparency and therefore prevents patients from price shopping for non-emergency medical services. While there is often an unusually large price variation across providers, consumers largely have no way to predict how much they will pay for services due to an obfuscated pricing system, where insurance companies negotiate prices with hospitals. According to a survey, majority of US patients support the idea of price shopping for medical services and do not believe that higher-cost providers are of higher quality, yet only 3% of polled people had compared costs across providers before receiving care. Most cite “difficulty obtaining price information” as a reason. [7]
And price variation there is! Within the same market of Philadelphia, the most and least expensive options were shown to vary 3x and 7x for knee replacement and limb MRI respectively. In other words, you can get an MRI for $400, or for 7 times as much ~3K depending on the hospital (see Figure 1). In a similar fashion, a metabolic blood panel test can cost $11 or $952 within the same city of Tampa. [6]
Quality could be argued to be a factor, but sometimes prices seem to be higher just because hospitals (sellers) know patients (buyers) either have no leverage or information to make price informed decisions. To support that hypothesis and general sentiment, let’s look at market analysis and drug pricing facts:
- Higher rates: hospitals in monopoly markets have prices that are 12% higher than those in markets with four or more providers, even after controlling for differences in cost and clinical quality. [yale]
- Identical drug pricing variation: recently, a hospital billed 16K/vial for an anti-venom drug that on average costs 3k/vial. [source]
- Insulin arbitrary inflation rate: Insulin is a life necessary (type 1 diabetes) drug for 0.55% , needed (type 2 diabetes) for 8.6% of US population. [cdc] Despite consistent cost of production, insulin has seen an inflation 1150% above overall economy inflation of 55.6% for the same time period (see Figure 2).
Cost transparency can foster competition between different hospitals and allow patients to choose the most cost-effective option for non-urgent care [1:Hastings College of the Law]. Recently, cost transparency has been mandated from hospitals (sellers) [4] and the information is now sprouting online: ascension, healthcarebluebook, fairhealthconsumer. However, that information is not yet sufficient because while hospitals are required to publish the requested cost for services, the amounts that medical providers actually agree to accept from insurance companies continue to remain closely held secrets [6].
Notably, even though cost transparency can foster competition, it would not affect monopolies or collusions [3] that may exist in parts of the market. Risk of collusion is a popular counterargument against healthcare market transparency. It is argued that by knowing the price of their competitors, hospitals may collude legally and raise prices in unison. [5]
Collusion and monopoly disruption are exactly what antitrust efforts are for, which is why “discouraging anticompetitive behavior” is recommended to supplement healthcare transparency [1:Hastings College of the Law].
Part 2: Crowd-sourcing transparency
The US medical market is an example of how the seller can release information to increase transparency. But what happens if that information is not enough or nonexistent to correct for unfair information power dynamic? Buyers can also aggregate their individual purchase/experience information to get equivalent transparency (or approaching equivalence if only part of buyers release their information).
In particular, I focus on the pattern of third parties injecting themselves between buyers and sellers in marketplaces (see Figure 3).
Below I summarize how such approach has positively affected particular markets.
US Census vs. efficient funding allocation
First of all, the idea of crowd-sourcing information is far from new. Public polling for both opinions and statistics is commonplace; and the first US Census was conducted as early as 1790 [source]. However, mostly such information has been used privately and not made easily accessible to the regular consumer.
But the approach, effect, and concerns remain the same. In the case of the US Census, by understanding what people are where, the government can more efficiently allocate budget and target certain initiatives (e.g. education funding for high family density areas).
Glassdoor’s pay confidentiality in labor market
By using sites like Glassdoor and levels.fyi, professionals can easily check the market rate for their role/experience and use that in their salary negotiation or job-hunting, bypassing pay confidentiality. Something that employers actively tried to prevent and what was impossible a few decades ago.
- Better Wages: In a 1992 study of US labor market by economists Richard Hofler and Kevin J. Murphy, it was found that “wages were roughly 10 percent lower in the early 1990s than they could’ve been in a world with costless job information.” In today’s U.S. economy, that would translate into employees loosing roughly $757 billion per year in their wages. [8]
- Equality of opportunity: A 2004 study from economists Yannis Ioannides and Linda Datcher Loury found gender and racial gaps in access to informal sources of job information, potentially disadvantaging low-income and minority households who lack informal professional networks. [8] Correcting for that inequality, asking about company culture is gaining popularity on sites like teamblind.
Thus, salary and opinion aggregator sites in labor market serve as great examples of how by organizing together, people can correct unfairness and inefficiency.
Reputation mechanisms in goods&services market.
A similar pattern of crowd-sourced public wisdom can be seen in online reputation systems. Services like Yelp, Foursquare, Google Maps became third-party tour-guides to help buyers navigate sellers of goods and services.
Despite imperfections of such rating systems, they filled a need for experience quality transparency for consumers.
Each added star on Yelp has been associated with a 5 to 9 percent increase in revenue. [Harvard business review]
Reputation became a major point of consideration for anybody who wants consumers to buy their products or use their services. But it is the idea of ratings that is especially interesting. In an ideal world, people are leaving reviews to essentially create a crowd-sourced “guide to life”, helping others navigate something that was once uncertain to themselves.
Part 3: Trustworthiness vs. Privacy in crowd-sourced information
However, as crowd sourced reputation and information systems become more integral, they also become targets for misuse. Particularly, individual anonymity and lack of accountability makes it easier to abuse the system.
Case and point, services for providing fake reviews and subscribers are currently flourishing.
Twitter “follower factory” was covered by nytimes, where a company — Devumi (still active) — made its business selling bot subscribers to Twitter users, and continues to offer services like “influencer marketing”; whether those influencers are real or not, clients can only assume. Some high profile figures were linked to using Devumi’s services.
Similarly, there exist review farms to boost your business’s online reputation. Such misuse from even a few bad actors makes it harder to trust these systems and reduces their overall “health”. Recall the outrage from the alleged Russian election meddling.
Sometimes, easy solutions are available. For example, making sure the user leaving a review has a verifiable experience with the business. As such, Foursquare and google maps prioritize your review if you were physically at the place, Amazon shows “verified purchase” on reviews. But that is harder to do in other markets, where information is either more sensitive or harder to verify.
Yet for online democracies to work, an individual should only be allowed to cast one vote and be accountable for that vote. Therefore, we need to solve the problem of allowing people to remain anonymous online yet enforce their uniqueness and legitimacy where needed. Using specific encryption keys [14], or building decentralized reputation [11, 12] and identity [13] systems have been proposed as potential solutions. The idea is to make your individual trustworthiness a resource to manage; whereas, currently you can make a new account in most systems and start “fresh”. It doesn’t have to go all “Nosedive” from Black Mirror, but discouraging explicitly malicious actors can go a long way to increase the health of the system.
Google’s New Initiative’s fact checking support is a related project in-tune with this sentiment:
To keep many of our online ecosystems functional, platforms will need to invest in their trustworthiness and reduce the burden of skepticism from consumers.
Conclusions
I described multiple markets where information transparency has already shown a positive effect and also shown that such transparency does not necessarily need to come from the seller’s side. There are many cases, where sellers have an interest in keeping their buyers in the dark. Modern technology allows buyers to aggregate transparency themselves where necessary; albeit, with a little help from the tech community.
This piece is meant to demonstrate the value of individual information and inpsire more organised collective action.
Collectively, people can fill in information gaps to help each other navigate uncertainty.
References
[1:Hastings College of the Law] https://repository.uchastings.edu/cgi/viewcontent.cgi?article=2099&context=faculty_scholarship — a comprehensive review of healthcare market specifics, and pricing policy suggestions.
“The most effective combinations of these solutions will include two elements: (1) one that targets provider leverage by discouraging anticompetitive behavior, and (2) another that mandates disclosure of price information.“
“ Although transparent prices would allow the Attorney General to use discovery to gather useful data to inform an antitrust suit, because of the uncertain outcomes of enacting price transparency alone in regions where providers or insurers have substantial leverage, the first step in this process should be using antitrust litigation to break down and discourage anticompetitive uses of market leverage.”
[2] https://healthcarepricingproject.org — data driven study of US healthcare. Paper summaries.
[3]https://wolfstreet.com/2014/09/16/how-hospitals-and-health-insurers-collude-to-keep-prices-high/
[4]https://www.nytimes.com/2019/07/29/health/hospital-rates-insurers-trump.html
[5] https://www.nytimes.com/2019/06/24/upshot/transparency-medical-prices-could-backfire.html
“Recent work by the Health Care Cost Institute, a group that pools and analyzes insurance claims, found that the highest price for a simple blood test could be 40 times the lowest price for the same test in a given market.”
[6] https://www.nytimes.com/2019/04/30/upshot/health-care-huge-price-discrepancies.html
[7] https://www.healthaffairs.org/doi/abs/10.1377/hlthaff.2016.1471?journalCode=hlthaff
[8] https://www.glassdoor.com/research/app/uploads/sites/2/2015/04/GD_Report_2-2.pdf
[9] https://www.cato-unbound.org/2015/04/06/alex-tabarrok-tyler-cowen/end-asymmetric-information — further reading on reputation mechanisms.
[10] N. Confessore, G. Dance, R. Harris, M. Hansen “Follower Factory”, https://www.nytimes.com/interactive/2018/01/27/technology/social-media-bots.html
[11] E. Androulaki, S. Geol Choi, S. Bellovin, T. Malkin, “Reputation Systems for Anonymous Networks”. https://link.springer.com/chapter/10.1007/978-3-540-70630-4_13
[13] https://medium.com/linum-labs/the-whos-who-of-decentralized-identity-systems-433b2dd9a195