by Robert Litan
While I was at Kauffman, the foundation’s largest monetary commitment to entrepreneurship research was the funding of the most comprehensive longitudinal data set of new firms ever assembled about U.S. firms. This Kauffman Firm Survey (KFS) followed 5,000 firms established in 2004 (the year the study was launched) for eight straight years, asking over 100 questions of their founders or top leaders. By doing this, the KFS enables researchers to test numerous hypotheses about the factors driving entrepreneurial success, although it too suffers from a form of survivorship bias, though one not as severe as in the Bhidé study. The reason is that many firms fail or merge, dropping out of the KFS database over time, which means that only the survivors report the most complete data histories. However, unlike the Bhidé study that deliberately chose 100 of the most successful enterprises as the basis for its analysis, the KFS sample includes surviving firms that exhibit various degrees of success. Numerous researchers have made use of the KFS to author papers, including those of one of Kauffman’s own senior fellows, Alicia Robb.
As of this writing, however, by far the best and most recent empirically based study of and guide for potential and actual entrepreneurs in my opinion is The Founder’s Dilemmas, by Noam Wasserman of the Harvard Business School.27 Wasserman’s exhaustive treatment of virtually all aspects of the early stages of launching a new business is based on the experiences of nearly 10,000 founders, almost 20,000 executives, in about 3,600 startups over a 10-year period from 2000 to 2009. The book is chock full of advice, backed by an extensive analysis of the data on a wide variety of topics, including: whether to found companies alone or in teams; the issues that come up in teams that tend to lead to success or failure of the enterprise; the often tension-filled discussions about how to split up equity and compensate founders and their employees; hiring issues; and when to stay with the company (and if so, how to transition toward founder and management succession), or to sell out to a strategic buyer. This brief summary just scratches the surface of an enormously impressive book, which I believe is must-reading for anyone thinking about or in the process of founding a business. The book and its research provides clear evidence that academics can provide practical advice, provided they go out, like Wasserman, and talk to real people doing the things that entrepreneurs care about and that academics want to analyze.
The Bottom Line
Experiments are a part of everyday life and of many disciplines and pursuits, including both economics and business. Yet the movements toward experimentation in both lines of endeavor have developed, at least up to now, quite independently of one another.
This eventually will change. Both economists and those actively engaged in business now use experiments in laboratory settings and in the field. As this continues, I expect that businesses will call upon economists more frequently for their advice and their research lessons. Likewise, if economists are asked more frequently for their advice, expect to see them produce more business-relevant research.
Of the firms engaged in experimentation, perhaps the most important to the overall economy are new ones (speaking only of the United States), since they drive innovation, especially disruptive innovation, which determines the pace of economic growth. We may not need economists to start new businesses, but the better their research about what accounts for successful entrepreneurship, the larger should be the number of successful entrepreneurs and their companies. Or so we should all hope.
Notes
1. Vernon L. Smith, Memoirs (Bloomington, IN: Author House, 2008).
2. Megan McCardle, The Upside of Down: Why Failing Well Is the Key to Success (New York: Viking Adult, 2014).
3. This profile draws heavily on Smith’s memoirs cited in endnote 1.
4. Author’s e-mail correspondence with Vernon Smith, June 6, 2013, and Stephen Rassenti, Vernon L. Smith, and Bart J. Wilson, “Using Experiments to Inform the Privatization/Deregulation Movement in Electricity,” Cato Journal 21, no. 3 (Winter 2002): 515–544.
5. Daniel S. Hammermesh, “Six Decades of Top Economics Publishing: Who and How?” Journal of Economic Perspectives 51, no.1 (2013): 152–162.
6. The findings in this study were subsequently corroborated and refined in Daniel S. Hammermesh, “Six Decades of Top Economics Publishing: Who and How?” Journal of Economic Literature LI, no. 1 (March 2013): 162–172.
7. Much of their research is summarized in Abhijit Banerjee and Esther Duflo, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty (Cambridge, MA: Public Affairs, 2013). See also What Works in Development? Thinking Big and Thinking Small, ed. Jessica Cohen and William Easterly (Washington, DC: Brookings Institution Press, 2010).
8. James Manzi, Uncontrolled: The Surprising Payoff of Trial-and-Error for Business, Politics, and Society (New York: Basic Books, 2012), 195.
9. Tyler Cowen, Average Is Over: Powering America beyond the Age of Stagnation (New York: Dutton Books, 2013), 225–228.
10. For another discussion of randomized controlled experiments in economics see “Random Harvest,” The Economist, December 14, 2103, www.economist.com/news/finance-and-economics/21591573-once-treated-scorn-randomised-control-trials-are-coming-age-random-harvest.
11. Manzi, Uncontrolled.
12. Information for this paragraph in part from Brian Christian, “The A/B Test: Inside the Technology That’s Changing the Rules of Business,” Wired, April 25, 2012, www.wired.com/business/2012/04/ff_abtesting.
13. Eric Ries, The Lean Startup (New York: Crown Business, 2011).
14. Manzi, Uncontrolled, 147.
15. This profile is based on Manzi, Uncontrolled, and an interview with him, October 9, 2013.
16. See Amar Bhidé, The Origin and Evolution of New Business (New York: Oxford University Press, 2000).
17. William Baumol, Robert E. Litan, and Carl Schramm, Good Capitalism, Bad Capitalism, and the Economics of Growth and Prosperity (New Haven, CT: Yale University Press, 2007).
18. Robert E. Litan and Carl Schramm, Better Capitalism: Renewing the Entrepreneurial Strength of the American Economy (New Haven, CT: Yale University Press, 2012).
19. Edmund S. Phelps, Mass Flourishing: How Grassroots Innovation Created Jobs, Challenge and Change (Princeton, NJ: Princeton University Press, 2013).
20. The Economist magazine devoted an entire special report in its January 18, 2014, edition called “A Cambrian Moment,” about experimentation as the key to tech startups in particular.
21. Jeffrey Adelman, Worldly Philosopher (Princeton, NJ: Princeton University Press, 2013).
22. Albert O. Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States (Cambridge, MA: Harvard University Press, 1970).
23. Malcolm Gladwell, “The Gift of Doubt,” The New Yorker (June 24, 2013): 74–79. The book about Hirschman is Adelman, Worldly Philosopher.
24. Scott Adams, How to Fail at Almost Everything and Still Win Big (New York: Penguin, 2013). This is also one of the main themes of McCardle’s excellent book, The Upside of Down.
25. Norm Brodsky and Bo Burlingame, Street Smarts: An All Purpose Guide for Entrepreneurs (New York: Penguin, 2010).
26. Bhidé, Origin and Evolution of New Business.
27. Noam Wasserman, The Founder’s Dilemmas: Anticipating and Avoiding the Pitfalls That Can Sink a Startup (Princeton, NJ: Princeton University Press, 2012).
Chapter 7
Matchmaker, Matchmaker
At a fundamental level, economics is about resource allocation, or who gets what? Mainstream introductory textbooks suggest that prices play a critical role in the answer. Those willing and able to pay the going price for goods and services sold in the market get what suppliers make available. In competitive markets, this outcome is said to be efficient.1
However, there are many examples of markets where price is not the governing factor in decisions, if prices exist at all. Consider some of the most important decisions people make: where to go to college, which job to choose, or which person to
marry. Price is important in two of these cases, but not in the third, and even in the first two, price is often or generally not the most important factor in the decision.2
So how do these nontraditional markets work, where matches may be made based on a certain set of rules and characteristics that are unique to each setting?3 Answering that question is the subject of this chapter, in which I introduce you to a new branch of economics called market design, a field devoted to making markets work more efficiently by better matching supply with demand.
I will begin with a brief and gentle introduction to the economics of matchmaking. I then discuss two major applications of market design and matching theory in the real world, including one where business is clearly at stake (the job market) and the other where the Internet has enabled much larger businesses to develop, such as the dating market. I conclude with some bottom line observations.
Unlike the subject of the last chapter, where economists and businesses using experiments developed more or less in parallel, there has been considerable cross-fertilization among economists and businesses in the world of matchmaking. Given the stories you are about to read, I wouldn’t be surprised to see a lot more such cooperation in the years ahead.
A Gentle Introduction to Market Design and Matching Theory
Matching is one of the most fundamental functions of markets. As we have seen in previous chapters, prices are the matchmakers in traditional markets. Indeed, Adam Smith, the father of modern economics, marveled at the way consumers and firms operated in a market economy and responded to prices, acting as if guided by an invisible hand that leads markets to allocate resources efficiently. All of this you will find in any standard economics textbook.
A new field of economics, not yet found in most introductory or even intermediate-level economics textbooks, is developing. That field is market design, which recognizes that well-functioning markets depend on detailed rules. For example, supply and demand drive both the housing market and the job market, but someone who wants to buy or sell a house goes through similar but also some very different steps than those taken by job seekers or employers.4 Market designers try to understand the specific rules, characteristics, and culture of each market in order to fix them when they’re broken, or to build markets from scratch when they’re missing.5
Think of market designers as the engineers of economics. Just as civil engineers apply principles of physics and mechanics to design bridges, market designers apply the principles of economic analysis—competition, incentives, information, economies of scale—to design exchange mechanisms or improve existing markets.6
Matching theory essentially comes from two subfields of economics: game theory—the study of strategic behavior, in which particular rules and market characteristics play an important role in an outcome—and experimental economics, which is about conducting empirical work and testing theories in the real world.
To understand the field of market design and the economics of matchmaking, remember the discussion in Chapter 2 about how traditional markets can fail, or why markets for certain goods or services are sometimes missing in the first place: where there are externalities, information asymmetries (one party knows more than the other about a particular product or service), or public goods. A classic example is pollution, which imposes costs on third parties and therefore leads to overproduction when the producer does not internalize these environmental costs. Sometimes, the private market can solve these problems (if the polluters and those affected by it are small in number and thus can negotiate a solution), and in other cases, the government is better suited to provide remedies, which include taxes, subsidies, and regulation.
In the context of market design, market failure refers primarily to the specific rules (or lack thereof) and characteristics of markets that sometimes impede efficient matching. In particular, there are three kinds of market failure that the market designers study:
Markets can fail to provide thickness, or to bring together enough buyers and sellers to transact with each other to make a real market.7
Markets can fail to overcome congestion, or to provide participants with enough time, or with the means to conduct transactions fast enough to make satisfactory choices when faced with many alternatives.8
Markets can fail to make it safe for participants to reveal or act on confidential information they may hold.9
The primary motive for economists interested in improving market design is to correct these market failures, and more broadly, to make markets work more efficiently by better matching individual suppliers with purchasers or those on the demand side of transactions. As such, market designers are interested in improving both traditional markets governed by a price mechanism and nontraditional markets, where the scarce goods to be allocated are heterogeneous and indivisible and prices may not be key to reconciling supply and demand. Examples of the latter include which students go to certain schools, which workers get specific jobs, and who receives which transplantable organ.10
Many of the examples we will explore in this chapter involve nontraditional markets, but it is also useful to briefly consider the ways in which market designers can improve traditional markets that are governed by a price mechanism. Take auctions and online marketplaces, discussed in Chapter 3, for example. While in theory it seems straightforward that buyers and sellers in these settings would arrive at an efficient outcome, in practice it is actually very difficult to present buyers and sellers with only the markets and products or services they are interested in. That is, while it is easy to provide a platform to draw participants to a marketplace, there are numerous inefficiencies that arise when buyers and sellers cannot find the relevant products or services, either due to information asymmetries, congestion, or a lack of thickness. In these situations, market makers can improve price-driven markets by better matching supply and demand in accordance with the rules and institutional culture of each market.
Two of the pioneers of market design and matching theory are Lloyd Shapley and Alvin Roth, who were awarded the Nobel Prize in Economics in 2012. The Nobel committee recognized Shapley’s and Roth’s extensive work across a multitude of disciplines and settings, from abstract theory developed in the 1950s and 1960s by Shapley on stable matching to the empirical and practical work conducted by Roth since the early 1980s (see the following boxes for brief overviews of the work of each economist).
Lloyd Shapley
Lloyd Shapley, like a number of Nobel Prize winners in economics, did not get his formal training in the subject. Nonetheless, during the course of his career, he made significant contributions to game theory, market design, and matching theory, and was recognized by the Nobel committee in 2012 (along with Alvin Roth) for the importance of this work.
Shapley was born in Cambridge, Massachusetts, in 1923 and enrolled at Harvard University in the early 1940s, but couldn’t complete his degree, because like many other young men at the time, he was drafted and served in the U.S. Army from 1943 to 1945. Shapley returned to Harvard after the war and graduated with a degree in mathematics in 1948. He then worked as a research mathematician at the RAND Corporation before going to Princeton University to obtain his PhD in mathematics. While there, he mentored and befriended John Nash, a fellow future Nobel laureate, and the protagonist in the book and movie A Beautiful Mind. Shapley can even be credited with that title when he described Nash as having a “keen, beautiful, and logical mind.”11
At Princeton, Shapley’s major contribution to game theory was his introduction of what has come to be known as a Shapley value—a payoff derived from a set of axioms applicable to every cooperative game. The value and related measures of it have been widely applied in numerous settings over time, including in the quantification of the impact of voting rules on the influence of individual voters, in legal decisions concerning electoral districting and representation, and in accounting problems of cost allocation.12
Shapley also made some of the earliest and most important theoretical
contributions in matching theory. In a seminar paper he coauthored in 1962 with David Gale, Shapley explored the idea of stable matching—allocations where no individuals perceive any gains from further trade.13 In the paper, titled “College Admissions and the Stability of Marriage,” Shapley and Gale presented a model of a two-sided matching in which men and women—or students and colleges—expressed preferences for their matches. In particular, Shapley and Gale proposed a deferred-acceptance algorithm (since known as the Gale-Shapley algorithm) for finding stable matching, for instance, where no couples would break up and form new matches that would make them better off.
Since 1981, Shapley has been affiliated with the University of California, Los Angeles (UCLA), where he is currently a professor emeritus.
One of the primary distinctions market designers make when analyzing markets is that of centralization versus decentralization. Some markets have centralized clearinghouses that match supply and demand (even if prices are absent). Other markets are decentralized and do not have a central authority to allocate resources.
For example, consider the market for medical residents, with hospitals on the demand side and new doctors (residents) on the supply side. In the early twentieth century, medical students would apply for positions in hospitals during their final year of medical school. However, as competition for new doctors increased, by the 1940s hospitals began hiring students much earlier than before—often almost two years before graduation.14 Due to the uncertainty and insufficient information about medical students’ plans of where they wanted to work two years later, as well as the lack of viable alternatives that were available to each student, the market began to lose thickness over time.