Tipping the scales: Balancing consumer arbitration cases
- Arbitration is often assumed to be fair and balanced. But brokerage firms often hold an advantage over consumers.
- Random selection of arbitrators would increase consumer awards by about $60,000 on average.
- Industry-friendly arbitrators are 50 percent more likely to be chosen from the arbitrator pool than their consumer-friendly counterparts because securities firms are sophisticated at eliminating consumer-friendly arbitrators.
Arbitration is often assumed to be a fair and balanced — not to mention cheaper — alternative to civil lawsuits when it comes to adjudicating disputes between companies and either consumers or employees. But our detailed analysis shows that, at least when it comes to consumers and the securities industry, brokerage firms routinely hold a decisive and measurably unfair advantage over consumers.
Using data from consumer arbitration cases in the securities industry over the past two decades, my coauthors Mark Egan, Gregor Matvos, and I found that securities firms hold information and selection advantages over consumers that result in more industry-friendly arbitration outcomes. And we’ve been able to measure the results of this pro-industry tilt in monetary terms (Egan, Matvos, and Seru, 2021).
The advantages that securities firms have are the result of both an information advantage in selecting arbitrators as well as a pro-industry bias in the arbitration pool.
The information advantage comes because securities firms carefully track the results of arbitration proceedings over time. They especially note those arbitration firms that tend to deliver more favorable results to the securities firm — that is, lower compensation awards for consumers. These brokerage firms keep “strike lists” of arbitrators who are more “consumer friendly,” allowing them to strike out those arbitrators from the selection process. Tracking arbitration outcomes gives securities firms an information advantage over consumers when it comes to selecting an arbitrator, especially since consumers tend not to be as well informed about the track records of arbitration firms.
In part, this information advantage helps lead to a pro-industry bias in the pool. That’s because arbitrators, who of course are compensated only if they are selected, compete with each other to be chosen by securities firms. It may be that arbitration firms know they have a better chance of being selected and paid if they exhibit a more pro-industry stance. Or it may be that those arbitration firms that do exhibit a more pro-industry stance are more successful and thus tend to dominate the pool. Either way, there is de facto competition.
The information and selection advantages securities firms have are substantial. We calculate that if arbitrators were selected randomly without the input from either party, consumer awards would increase by about $60,000 on average, compared with the current system. Approximately 60 percent of that effect comes because firms are better than consumers at striking arbitrators from a given arbitrator pool. We calculate that competition among arbitrators accounts for the remaining 40 percent.
Ironically, we found that policies that are intended to benefit consumers — such as increasing arbitrator compensation or giving parties more choice — would benefit informed consumers but hurt the uninformed.
The rules of arbitration
When consumers purchase a product or service, the purchase agreement often contains an arbitration provision, which prohibits the consumer from suing the seller in court. Instead, it mandates that any dispute be resolved using arbitration. The parties present their case to arbitrators who then issue a legally binding dispute resolution.
Arbitration clauses have become nearly ubiquitous in American business. They are used by all brokerage firms, the largest insurance companies (AIG, Aetna, Inc., Blue Cross and Blue Shield, Travelers), the biggest banks and other financial firms (American Express, Bank of America, Chase Bank, Citigroup), as well as the largest fintech firms (PayPal, Venmo, and Square).
Arbitration clauses are also pervasive among non-financial firms such as online retailers (Amazon, eBay, Walmart.com), music service providers (Apple, Spotify), wireless providers (Verizon, AT&T, T-Mobile, Sprint), and sharing economy firms (Uber, Lyft, Airbnb), covering trillions of dollars of transactions.
The use of arbitration in recent years has grown to such an extent that far more disputes between consumers and companies in the U.S. are settled through mandatory arbitration than through the court system. The American Arbitration Association, the world’s largest provider of arbitration and alternative dispute resolution (ADR) services, reported overseeing 68,000 consumer cases in 2020. That compares with 12,922 consumer protection lawsuits filed in 2020.
A central feature of arbitration, and one that is ostensibly designed to ensure fairness, is the ability of both companies and consumers to exert some control over the selection of the arbitrator. In securities arbitration, each party is presented with a randomly generated list of arbitrators and both sides can then strike a limited number of arbitrators from the list. The list then gets whittled down to two or three arbitrators who handle the case.
The ability to select an arbitrator is a key feature because choosing an arbitrator can significantly affect the case outcome: “The selection of an appropriate arbitrator or arbitration tribunal is nearly always the single most important choice confronting parties in arbitration” (Stipanowich et al., 2010).
Consumer arbitration: A different animal
In the study of arbitration, the existing literature has mainly focused on situations in which both parties are equally informed, such as arbitrations between unions and employers, or arbitration in an experimental setting.
But our research focuses on consumer arbitration, specifically with securities brokerage firms. We examine the impact of the arbitrator selection process on consumer outcomes, a situation in which securities firms hold a considerable informational advantage in the key process of selecting arbitrators.
Our research had two goals. The first was to determine and establish several motivating facts that suggest that firms hold an informational advantage over consumers in selecting arbitrators, resulting in industry-friendly arbitration outcomes. The second goal was to develop and calibrate a model of arbitrator selection in which firms hold an informational advantage in selecting arbitrators. The reason for creating the model was to allow us to separate out and distinguish the effects of the two advantages that securities firms have in arbitration, namely: 1) The ability to choose pro-industry arbitrators from a given pool and 2) The pro-industry tilt in the arbitration pool that arises because of competition between arbitrators.
The model reveals that accounting for these advantages is critical in assessing any proposed changes to arbitration design. For example, policies that would benefit consumers, assuming they were informed, would actually hurt them if they were not, illustrating the importance of investor sophistication in consumer financial markets (Agarwal, Chomsisengphet, Mahoney, and Stroebel, 2015; Argyle, Palmer, and Nadauld, 2019; Anderson, Campbell, Nielsen, and Ramadorai, 2020; Goldsmith-Pinkham and Shue, 2020).
The model also allowed us to calibrate consumer gains or losses when arbitrator selection rules are changed, such as those in recent policy proposals. This allows us to speak to market design in financial markets, by showing that increased competition is not always beneficial (Budish, Cramton, and Shim, 2015; Budish, Lee, and Shim, 2019; Aquilina, Budish, and O’Neil, 2020).
Examining the securities industry
Arbitration in the brokerage industry is interesting in itself. Roughly 20 million U.S. households hold brokerage accounts, amounting to $20 trillion in assets (2016 Survey of Consumer Finances). The cases involve significant amounts of money. The mean and median damages requested by consumers are $760,000 and $240,000 respectively, providing substantial incentives for the parties in arbitration.
We studied arbitration in the securities industry using a new data set of roughly 5,000 disputes between consumers and financial advisers over the period 1998 to 2019. Our data on securities arbitration comes from the Financial Industry Regulatory Authority (FINRA) Arbitration Awards Database, which we merge with FINRA’s BrokerCheck data using unique case-level identifiers. The merged data allows us to observe detailed information on the consumer claimant, the securities firm respondent, the arbitrators, as well as dispute details and awards.
All disputes are resolved under the auspices of FINRA, which provides a uniform pool of arbitrators, as well as rules governing arbitration. The selection system used by FINRA is similar to that of the largest consumer arbitration forums such as the American Arbitration Association (AAA) and Judicial Arbitration and Mediation Services, Inc. (JAMS).
Most important for the research design, FINRA randomizes the list of potential arbitrators from which the parties select the arbitration tribunal, which we exploit. For a significant subset of cases, we also observe the randomly generated list of potential arbitrators presented to both parties, in addition to the arbitrators who were selected to the case.
FINRA established the Dispute Resolution Task Force in 2014 to investigate concerns that the arbitration procedures lead to outcomes favoring the industry. More recently, the Consumer Financial Protection Bureau (CFPB) proposed a new rule regulating mandatory arbitration clauses in certain financial products (Arbitration Agreements, 12 C.F.R. § 1040, 2017). Understanding arbitration design in the financial industry, therefore, has direct policy relevance.
We demonstrated in two steps that firms have a distinct informational advantage in selecting arbitrators. Later, we used these facts to establish the motivating assumptions behind our quantitative model.
First, we confirmed the intuitive belief by many practitioners that some arbitrators are systematically more industry friendly and that others are more consumer friendly (Stipanowich et al., 2010). But how big of an effect is this slant or bias? We found controlling for industry-friendly arbitrators alone as a factor accounted for 36 percent of the variation in awards we studied. An arbitrator who is more industry friendly by one standard deviation awards 14 percentage points (pp) smaller damages relative to the damages requested. For a median case ($240,000), this translates to a $33,600 smaller award for the consumer.
It is therefore not surprising that, anecdotally, brokerage firms maintain proprietary internal arbitrator rankings, or arbitrator “strike lists,” to guide their arbitrator selection process. Arbitration rules allow both parties a certain number of “strikes” to eliminate arbitrators from a randomly selected pool from which they must eventually choose.
Second, we find that securities firms have an informational advantage over consumers in choosing arbitrators who are favorable to them. The list from which the parties select arbitrators is randomly generated by FINRA. If both parties were equally well informed, they would strike arbitrators who favor the opposing side and the median arbitrator from the list would be chosen. Under that assumption, being industry or consumer friendly should not increase selection chances.
But, instead, we find that industry-friendly arbitrators are 50 percent more likely to be chosen from the list than their consumer-friendly counterparts (see Figure 1).
Figure 1: Are Industry-Friendly Arbitrators Selected More Frequently?
Several tests suggest that these patterns are not due to unobserved case characteristics driving arbitrator selection. Rather, the strongest evidence indicates that arbitrator selection arises because firms are better at eliminating consumer-friendly arbitrators.
Additional evidence reveals the advantage securities firms have is driven by their extensive experience with arbitration. This informational advantage by securities firms, we found, is substantially reduced when consumers are better informed, such as when they use an attorney who specializes in arbitration and is a member of the Public Investors Arbitration Bar Association (PIABA).
What our model showed
After examining the distinct informational advantage in selecting arbitrators securities firms had, we used these facts to establish the motivating assumptions for a quantitative model of arbitrator selection.
The model highlights a second advantage of the informed party: Because arbitrators compete to be selected by the informed party (the firm), the whole pool of arbitrators becomes more industry friendly in equilibrium. This competition effect can be large, because selection is determined by how industry friendly arbitrators are relative to other arbitrators.
We use the calibrated model to break down the equilibrium advantage of the informed party into these two components — that is, the information advantage and the biased pool advantage. We then use the model to show how accounting for the informational advantage is critical when it comes to assessing changes to arbitration design. If the informational advantage of firms is ignored, several policy changes that aim to help consumers actually hurt them instead. We use the calibrated model to evaluate the magnitude of consumer gains and losses across different proposals.
In practice, securities firms and consumers strike arbitrators from a randomly generated list. Arbitrators can naturally differ in their underlying beliefs about what constitutes a fair award. But they also can depart from these beliefs and choose how consumer or industry friendly they are, i.e., their “slant.” This concept of slant is similar to the choice of political slant in the media industry (Gentzkow and Shapiro, 2006) and judicial discretion in court decisions (Gennaioli and Rossi, 2019).
Arbitrators are compensated only if they are selected to arbitrate a case. They compete with other arbitrators to be selected on the arbitration panel. In doing so, they can and do trade off their preferences for a fair award in exchange for monetary compensation from arbitration. Sophisticated securities firms observe arbitrators’ slant; consumers, on the other hand, are generally uninformed about the biases.
A key result of the model is the conclusion that, because arbitrators compete to be selected, the whole pool of arbitrators becomes industry friendly, increasing the informational advantage of firms. Why wouldn’t this be balanced by consumer-friendly firms competing? Because consumers generally do not have an information advantage in choosing arbitrators, as they have not examined the records of the arbitrators in the pool.
Even though the underlying beliefs of arbitrators may be unbiased, competition among arbitrators drives all arbitrators to intentionally slant their case decisions in favor of firms.
This competition among arbitrators exacerbates the informational advantage of firms. This effect can best be seen in the special case when arbitrators have no concerns about fairness and only want to maximize their monetary payoffs. In that situation, every arbitrator wants to be a bit more industry friendly than other arbitrators. This way, firms will not strike them from the list, increasing the arbitrators’ chances of earning arbitration fees. Because all arbitrators want to do the same, this results in a “race to the bottom.”
Our counterfactuals suggest that re-designing incentive compensation and arbitrator selection design can ameliorate the pro-industry tilt, but only if the design accounts for uninformed consumers. We show examples of policies, such as increasing arbitrator compensation or giving parties more choice, benefit consumers if they are informed, but hurt them if they are uninformed.
One avenue for future research is to examine the extent to which this result is generic. More broadly, our findings suggest that limiting the firm’s and consumer’s inputs over the arbitrator selection process could significantly improve outcomes for consumers. We hope that future work can extend our workhorse quantitative model to evaluate such alternative arbitration selection mechanisms.
Amit Seru is a SIEPR Senior Fellow, a Professor of Finance at the Stanford Graduate School of Business, and a Senior Fellow at the Hoover Institution. His primary research interest is in corporate finance. He is interested in issues related to financial intermediation and regulation, interaction of internal organization of firms with financing and investment, and incentive provision in firms.
Aquilina, M., E. Budish, and P. O’Neill. “Quantifying the High-Frequency Trading ’Arms Race’: A Simple New Methodology and Estimates.” Working Paper.
Agarwal, S., S. Chomsisengphet, N. Mahoney, and J. Stroebel. 2015. “Regulating Consumer Financial Products: Evidence from Credit Cards.” The Quarterly Journal of Economics, 130 (1): 111–164,
Andersen, S., J.Y. Campbell, K.M. Nielsen, and T. Ramadorai. 2020. "Sources of Inaction in Household Finance: Evidence from the Danish Mortgage Market." American Economic Review, Forthcoming.
Argyle, B., C. Palmer, and T. Nadauld. 2019.” Monthly Payment Targeting and the Demand for Maturity. Review of Financial Studies, Forthcoming. Ashenfelter, O. 1987. “Arbitrator behavior.” The American Economic Review P&P, 77(2): 342-346.
Arbitration Agreements 12 C.F.R. part 1040 2017.
Budish, E., P. Crampton, and J. Shim. 2015. “The High-Frequency Trading Arms Race: Frequent Batch Auctions as a Market Design Response.” The Quarterly Journal of Economics, 130(4): 1547–1621.
Budish, E., R. Lee, and J. Shim. 2019. “A Theory of Stock Exchange Competition and Innovation: Will the Market Fix the Market?” National Bureau of Economic Research. NBER Working Paper No. 25855.
Egan, Mark L., G. Matvos, and A. Seru. 2021. “Arbitration with Uninformed Consumers” National Bureau of Economic Research, NBER Working Paper No. 25150
Goldsmith-Pinkham, P., and K. Shue. 2020. “The Gender Gap in Housing Returns.” Journal of Finance, Forthcoming.
Stipanowich, T., C. von Kannn, and D. Rothman. 2010. “Protocols for Expeditious, Cost-Effective Commercial Arbitration: Key Action Steps for Business Users, Counsel, Arbitrators & Arbitration Provider Institutions.” The College of Commercial Arbitrators.
2016 Survey of Consumer Finances. 2016. Federal Reserve Board.