—Francis Bacon, English philosopher, statesman, and scientist

THROUGHOUT HUMAN HISTORY, THE TWO MAIN FORMS OF economic interaction have been resource-grabbing and mutual cooperation. Resource-grabbing, which appeals to our selfish side, helps enlarge the slice of the societal pie that a person receives, but it often reduces the size of the pie itself. Mutual cooperation, which appeals to our empathetic and compassionate side, increases the size of the pie but not necessarily of the slice that each person receives.

Adam Smith had the brilliant insight that in competitive markets, mutually beneficial cooperation could be sustained by relying on the self-interested component of human nature. He didn’t think that people lack a compassionate side; indeed, he recognized its existence at the beginning of The Theory of Moral Sentiments. Nor did he think that people should be selfish. Rather, he thought that if a system of cooperation could function even if people are selfish, it must be robust and resilient.

As Smith correctly pointed out, and as I have emphasized throughout this book, the genius of the free-enterprise system is competition. It is competition—the famous invisible hand—that aligns individual incentives (the selfish desire to have more) with social incentives (the desire to increase the size of the economic pie).

It took economists 150 years to formalize in mathematical models Smith’s brilliant intuition in The Wealth of Nations. In doing so, they assumed the existence of well-functioning institutions. Yet markets do not arise spontaneously; they are human constructs and depend on a functional legal framework. The design of this framework and its enforcement affect how markets function and how the surpluses they generate are distributed. Besides the broader framework of law, markets also need very specific, smaller-scale rules to operate. Even the most competitive of all markets, the organized exchanges for commodities and securities, are shaped by hundreds of such rules.

From a distance, the rules may not seem that important. But they matter a great deal for the profitability of the parties involved in exchanges and for the efficiency of the market itself. A rule as trivial as the one specifying the number of digits in which securities are quoted has important effects. Securities in the United States were traditionally traded in fractions of eight, a convenient practice in the pre-computer world. In 2001, the Securities and Exchange Commission mandated a move toward the decimalization of quotes. This reduced the bid-ask spreads by 37 percent for securities listed on the New York Stock Exchange and by 50 percent for those traded on NASDAQ. It also reduced the amount of stock a trader was willing to buy or sell at the posted quote by 60 percent.¹ While the first outcome favored the small traders, who saw their transaction costs diminish, the second one potentially penalized the big ones, who saw the market’s depth reduced.

The same tension between resource-grabbing and cooperation that exists in the economic sphere is present in the political sphere, where the rules of the game are designed. Does Smith’s invisible hand also work here?

The feature necessary for the invisible hand to work in the economic realm is that individuals’ decisions influence other individuals only through prices, not directly. Even if everybody else chooses to buy a red tie, I am free to purchase a blue one. The only way decisions by others affect me is through the price of red ties (which will go up). This rule clearly does not apply in the political arena. When politicians pass a law requiring securities to be traded in dollars and cents, rather than in fractions of eight, they have a direct impact on my ability to trade them.

The invisible hand can, however, shape the design of rules if the institutions designing the rules compete among themselves. For example, US states compete with one another for business and population, which could lead them to design the most efficient laws—not necessarily the most advantageous ones for powerful incumbents. Similarly, to the extent that rules and regulations affect the quality and price of traded goods and services, international free trade creates pressure toward more efficient rules and regulation and away from crony capitalism, a point Raghuram Rajan and I made in our book Saving Capitalism from the Capitalists.²

The problem is that most established institutions wield market power. The competition among organized exchanges is far from perfect, for example, and that among countries is even farther. While some rules (such as a restriction on capital movement) have direct and immediate negative effects on international competitiveness, other rules take time, sometimes decades, to make their damage manifest; in the meantime, incumbents can profit from market distortions. Further, pressure from foreign competition disciplines small countries far more effectively than it disciplines a large, influential country like the United States. There is no guarantee, therefore, that the political process will generate rules for the economy that increase the size of the pie instead of benefiting politically powerful groups. This is where the free-enterprise system can degenerate into crony capitalism.

Most free-market supporters think that the best way to prevent the rules being designed to favor powerful incumbents is pure laissez-faire. But pure laissez-faire is not lack of government intervention; it is active intervention to protect the status quo. (French philosopher Jean-Paul Sartre illuminated the fact that we cannot avoid choices: not choosing is itself a choice.) If the interests of incumbents coincide with market development, as has occurred several times in history, laissez-faire does indeed lead to competition and free markets. Yet if it does not, pure laissez-faire can breed crony capitalism. Meanwhile, massive government intervention, as we have seen throughout this book, is plagued with problems and leads to worse forms of cronyism. So what is the solution?

In my view, government intervention in the economy should be so limited and so simple that voters can understand it, monitor it, and, if necessary, oppose it. What ultimately makes a capitalism for the people possible is a government by the people. Democracy is as essential to free markets as free markets are essential to democracy. Without free markets, freedom of speech and of the press becomes empty. Without democracy, the possibility of having truly free and competitive markets vanishes.

While democratic control of the rules of the economic game is essential, it is not without its own problems. As political scientist Anthony Downs explained more than fifty years ago, it is in voters’ interest to remain ignorant. Yet if they do remain ignorant, they can’t recognize policies that promote competitive markets—and politicians then have no incentive to make those markets more competitive.

Downs’s argument is basically sound, but it goes too far. It is neither feasible nor desirable for all citizens to become informed on all issues. Nevertheless, most people have some interest in public affairs and spend resources to cast informed votes. Some are willing to spend time and money to buy and read books like this one. While I tried to make this book as entertaining as its complex subject (and my ability) allowed, reading it requires time and effort. It offers no advice on how to make money, nor does it provide any self-help tips. Your reading it to the end is proof that you are concerned about public affairs and are willing to make an effort to improve our institutions. The question is how to leverage a limited amount of similar effort by many people as an effective way to contain cronyism.

Earlier, we saw how muckraking journalists helped improve American capitalism by exposing corruption in business and government. At the dawn of the twenty-first century, corruption is more subtle but no less dangerous. To identify and expose it, we need data and people—not muckrakers alone—to analyze those data rigorously.

THE POWER OF DATA

While there is some truth in Mark Twain’s celebrated statement that “there are three kinds of lies: lies, damned lies, and statistics,” it reflects an excessive mistrust of numbers. Certainly statistics can be used to lie or mislead. Liars, however, will lie with or without statistics. The more relevant question is whether data analysis makes it easier or harder to lie or, more broadly, to create unfounded arguments to support vested interests.

I believe that data, when subjected to rigorous cross-checking, make it more difficult to support unfounded arguments. Part of this conviction comes from two decades spent at the University of Chicago. The school boasts a strong tradition of empirical research and an even stronger tradition of lively (read: brutal) seminar discussions. At most universities, a speaker is accorded respect, especially when he is an established member of the profession. At Chicago, status elicits no such respect. Our new dean, recruited from Stanford University, decided to start by giving a seminar in his area (operations research). A young assistant professor attacked him, saying that it was “unfortunate” that he made certain assumptions. What in most other places would be considered lack of respect is at Chicago an expression of collegiality. The system isn’t always easy. Even after twenty years—I have to admit—I dread giving a seminar there. But I know that my work benefits tremendously from the sometimes painful experience. For this reason, we arrogantly say that a colleague loses twenty IQ points when he leaves Chicago: he misses the salubrious experience of the seminar. Over the years, I have come to compare this seminar with a colonoscopy: it is unpleasant but good for you.

The essence of rigorous data analysis is threefold: good data, smart people competing to find flaws in the analysis, and a level playing field where ideas matter and status doesn’t. This combination can produce amazing results. Bill Christie and Paul Schultz, the two researchers who identified the “collusive” quotes on NASDAQ, were PhDs from Chicago.³ So were the authors who identified the abnormal returns of mutual fund managers when they trade stocks of companies whose CEOs are college mates, a finding so sensitive that they had to avoid using any company names to escape being sued.⁴ More recently, one of my students at Chicago, using Russian bank transaction data, was able to estimate that 60 percent of Russian firms employ schemes to evade taxes by at least 40 percent.⁵ His findings were so detailed and explosive that I feared for his life, given Russia’s lawlessness. His seminar at the Russian Central Bank had to be canceled when the vice governor who invited him was assassinated. Using Illinois transaction data, another Chicago PhD student, working before the financial crisis began, found that 16 percent of the highly leveraged house purchases inflated the transaction price.⁶ For this to happen, both the lender and the borrower had to be complicit in the deception. I still remember the sense of disbelief that filled the room at the school when he tried to defend his ideas. Eventually he was vindicated. These researchers don’t see themselves as heroes dedicated to cleaning up the corporate world. They simply are pursuing their self-interest: to publish and to get tenure. Nevertheless, in the proper competitive environment their work helps identify dodges and swindles. Indeed, to limit crony capitalism, we need good data and powerful incentives to analyze them.

DATA TO THE PEOPLE

When I was a young assistant professor I worked with confidential data at the Bank of Italy. For supervisory purposes, the Bank of Italy collects all the information (including the amount and rate charged) on all loans above a minimum threshold made by banks to Italian clients. While each bank can observe the amounts other banks lend, the rates remain hidden; only the Bank of Italy can see them. These data are so confidential that I had to work with a bank employee and I could observe only summary statistics, not individual data points.

The first paper I wrote was fairly innocuous; I examined whether banks lower the rate they charge when a firm goes public. Grasping the power of these data, I became more ambitious and asked permission to use them for a different project: to test whether industrialists sitting on a bank’s board were getting a sweetheart deal from that bank. It is a simple question that could be answered with those data. It is also a question that the Bank of Italy, as supervisor and regulator of the Italian banking sector, should have found interesting. If the industrialists were getting sweetheart deals, it could weaken the bank and possibly jeopardize the stability of the banking sector. In spite of the potential regulatory implications and the brilliance of the idea (or so I thought), I was stonewalled. In my naïveté, I could not understand why bank officials seemed uninterested in finding out the truth. Now that I am older and more cynical, I think that the bank had no desire to confront reality. Its officials doubtless suspected the truth but wished to keep it hidden.

Data are a major challenge to power. All dictatorial regimes—including, notoriously, the Chinese government today—tightly control data and often manipulate them to prevent people from seeing the regimes’ shortcomings. Sometimes the regimes get so involved in the manipulation game that the dictators themselves end up being manipulated. Italy entered World War II partly because Mussolini thought that his army was in better shape than it actually was. And he thought so because he was a victim of his own regime’s propaganda.

Dictators are not the only ones who hide or manipulate data. CEOs try to do it all the time. Even honest ones have been known to use data to “beautify” results and hide their true performance. Democratically elected governments do it all the time, too. This effort to spin or control data is an indication of the disruptive power of information.

American banking regulators, for example, produce a confidential rating on bank solvency known as a CAMELS rating (for capital adequacy, asset quality, management and administration, earnings, liquidity, and sensitivity to market risk). The higher the CAMELS rating, the weaker a bank is considered. While there may be (debatable) reasons to keep these ratings confidential for a period of time (e.g., preventing a run on a weak bank), there is no persuasive reason to prevent their distribution after, say, a few years. Interestingly, though, the regulators aren’t keen to see the data circulate, since they also reveal the regulators’ own performance. A colleague of mine has tried to shed some light on the performance of bank regulators by teaming up with Fed employees with access to the data.⁷ Yet the Fed will ultimately decide what data can be released. So much for freedom of inquiry!

In the past, data collection, storage, and analysis were extremely costly. This cost provided a justification for the people in power to refuse to collect data and analyze them. Today, this justification is gone; in a digital world, data are collected automatically and are easy to store and analyze. To protect the analysis of data from bias, however, access to them and the release of whatever they show should not be conditioned on the findings. This is one of the most important and often ignored benefits of mandatory disclosure. “Publicity,” wrote Justice Louis Brandeis, “is justly commended as a remedy for social and industrial diseases.”⁸

In part, the greater availability of data is a side effect of the evolution of the Internet. Much has been written about the role of social networks in igniting the 2011 Arab Spring protests. Less has been said about the Internet’s role in making corporate data easily accessible. I have heard John R. Emshwiller, the Wall Street Journal reporter who uncovered the Enron scandal, credit his reporting to data available on the Internet. While waiting for an interview with Enron’s then-CEO Ken Lay, he looked up the company’s annual report online and noticed a strange footnote about special purpose vehicles, which can be ways firms hide debt off their books, away from investors’ eyes. Lay’s evasive answers on the subject led Emshwiller to dig deeper. The easy accessibility of the information proved crucial to exposing Enron’s deception. The amount of business information available on the Internet is amazing. I have even found circulating online “secret” internal memos of companies for which I serve as a board member.

Dictatorial regimes, like China, have tried to muzzle the Internet through the use of filters. Yet the design of the Internet makes this hard. Historically, the media have been dominated by means of communications that diffused information from a center to a periphery, from few to many. Books, radio, and TV fit this description. Information spread in this way is easy to control and manipulate. This is not true for the Internet.

Arpanet, the direct precursor of the Internet, was funded by the Pentagon. While Bob Taylor, the official in charge at the time, denies that Arpanet had a military purpose, its decentralized structure resembles a technology developed by a researcher at the RAND Corporation to permit long-distance communication in case of nuclear attack. The weakest point in any communication system is the coordinating center. If it goes down, the entire system fails. By contrast, the decentralized nature of the Internet makes it resilient to attacks. It is precisely this feature, whatever its origin, that makes filtering information difficult and allows communication from many to many.

Yet relying on the Internet alone to make business data available is not enough. From over-the-counter markets to private equity markets, from government regulators to corporations, there are plenty of data that can shed light on mistakes, inefficiency, corruption, and unethical behavior. While industrial secrets and privacy should be protected, there is no justification for restricting access to old data, such as old hedge fund trades or CAMELS ratings.

After the space shuttle Challenger disaster, the best minds of the time—including Nobel laureate physicist Richard Feynman—were employed in an investigation in which no stone was left unturned. Ultimately, the culprit was identified as a defective O-ring that became too stiff at low temperatures and caused a leak. To convince the public, Feynman demonstrated that conclusion with a televised experiment.

A serious act of data omission was committed by the Financial Crisis Inquiry Commission (FCIC), charged with finding the cause of the 2008 financial crisis. Economics might not be as precise a science as physics, but this cannot justify the failure of the commission, which investigated the financial crisis. With sufficient data, economists do have methods to identify the likely causes of an economic phenomenon. We are even better at refuting potential explanations. The availability of data might impose a limit on our ability, but not our methodologies. In the case of the financial crisis, the data were not made available to economists, because the interested parties were (and remain) afraid to share those data.

Unfortunately, the FCIC—composed mostly of elected officials, rather than experts—wasted its time in political squabbles. Its members could not agree even on how to define subprime mortgages and calculate how many such mortgages there were in the United States at the time of the crisis. A subsequent investigation into the work of the commission was more thorough than the commission’s actual investigation of the crisis. Details in the commission’s e-mails led the authors of a congressional report to find that the commission’s work was guided by politics rather than by facts. After all, the “fixes” to the crisis had already been decided and approved, before any fact was found, and the commission’s focus was on supporting or discrediting (depending on individual commissioners’ political party affiliation) the Dodd-Frank legislation rather than on establishing the truth. It was a great opportunity lost.

With its subpoena power, the commission could have collected and released the data needed to answer many crucial questions about the crisis. Did companies that compensated their traders (and not just their CEOs) more highly take more risk? Did financial institutions assume excessive risk out of incompetence or stupidity, or were they responding rationally to the implicit guarantee of bailouts offered by the government? Did the market see the spread of lax lending standards and price the relevant pools of loans accordingly, or was it fooled? Who were the ultimate buyers of toxic financial products, and why did they buy them? How important was the role played by fraud? These questions are likely to remain unanswered without disclosure of data. If we don’t answer them, there is a risk that we won’t find out what really caused the financial crisis until the next one occurs.

PEOPLE TO THE DATA

All the data in the world are useless, however, if researchers fail to analyze them thoroughly. A noted macroeconomist once declared that the real world exists to provide data input for his models. This was a joke, of course, but it contained an element of truth. All too often in seminars I have seen economists reject data in favor of their preferred theories, the opposite of the proper scientific method, which is to correct or reject theories in the light of new data. This reflects an ivory-tower attitude, in which the means justify the ends. Our discipline should serve the economy, not the other way around.

At the beginning of the twentieth century, American writers broke with the European tradition of literary journalism and created investigative journalism. Making this important change possible were a competitive newspaper industry and a vast market of educated voters who made investigative journalism profitable. This revolution would not have occurred, however, if writing for the masses had retained its reputation as a lowbrow activity, as it is still considered in many academic circles. At the beginning of the twenty-first century, we need a new kind of investigator—academic muckrakers. The conditions are ripe. The American academic market is the most competitive in the world and demand is not lacking. However, we need to overcome the (bad) academic social norm that frowns on academics writing for a wider audience.

CONCLUSIONS

Economists like to think about inefficiency in terms of missing markets. Markets provide an effective way to harness individual self-interest to produce the common good. Consequently, when such harnessing fails, we believe that a market must be missing. Pollution, for instance, can be seen as the effect of the absence of a market for pollution rights. Not surprisingly, economists have tried to address the pollution problem by creating the market deemed to be missing (i.e., a market for emissions permits).

I believe that the ultimate missing market is for the right to design the rules of the market itself. Society as a whole benefits from a competitive market, but nobody is willing to pay the cost of creating it. In principle, political competition should provide the missing incentive. After all, if the public benefits from a competitive market, the politicians defending this principle should triumph at the polls. Unfortunately, the political market is distorted by asymmetries in power and by voters’ ignorance on economic matters. If voters aren’t informed on what policies truly promote competition, it won’t pay for politicians to promote those policies. The distorted political market allows crony capitalism to triumph. We have seen that social norms can help us address market failures. I hope that this book helps create awareness of the need to accord social prestige to any effort to promote competition and resist crony capitalism. Academia can play a big role in this respect. If it fails to do so, it must be considered part of the problem.