September 18, 2011
The Nirvana Fallacy Revisited
The Nirvana Fallacy Revisited—Posner
I recall hearing when I first came to the University of Chicago in 1969 the expression "Nirvana Fallacy," used to describe the belief then dominant in the economics profession that market failures could and should be rectified by government intervention, assumed to be apolitical and effectively costless. The belief was unsound; government failure is commonplace, partly because of politics, partly because of the intrinsic difficulty of many of the tasks that are given to government to perform.
The opponents of the Nirvana Fallacy did not deny the existence of market failures; they just wanted the costs to be balanced against the cost of government intervention. But as the years went by there was a growing tendency among conservatives to regard markets as Nirvana—as self-regulating—and thus to deny the need for government regulation. Alan Greenspan, when he was chairman of the Federal Reserve Board, was a spokesman for this position. It became particularly influential during the administration of the second President Bush, with seriously adverse consequences. The deregulation of the banking industry, which had begun under President Carter and been completed during Clinton's second term, coupled with extraordinarily lax regulation of the nonbank banks (such as Goldman Sachs, Merrill Lynch, and Lehman Brothers) by the Securities and Exchange Commission (which had the principal regulatory authority over the nonbank banks) under the last chairman appointed by Bush, lax regulation of insurance companies (such as AIG) by state insurance commissioners, lax regulation of Fannie Mae and Freddie Mac by the Federal Housing Finance Agency, and lax enforcement by the Federal Reserve Board and the other bank regulatory agencies of the remaining regulations of commercial banks, were major causes (along with the lax monetary policy of the Federal Reserve in the early 2000s and misleading statements by successive Fed chairmen) of the financial crisis of September 2008 and the ensuing economic downturn—the most serious since the Great Depression.
Most economists did not understand the inherent instability of financial markets (which derives from the basic financial model of borrowing short term and lending long term, which can induce runs, especially when, as in the case of the nonbank banks, the short-term capital—often overnight—is not insured), the vulnerability of housing markets (in which the banking industry was heavily involved) to bubbles, or the potential macroeconomic consequences of a failure of those markets, which made deregulation a riskier policy than in industries such as air and surface transportation, electrical distribution, natural-gas production, oil pipelines, and communications. Because of the potential for catastrophic market failure, regulation should have been much tighter than it was.
Of course more than mistakes by economists were involved; the pressure of the banking industry for deregulation and light enforcement of the remaining regulations was intense, because the bankers wanted to be allowed to take more risk so that the expected return would be greater. Whether there would have been more resistance if the economics profession had opposed the industry is an academic question.
Regulation had got a worse name than it deserved because of a tendency to conflate it with other, more questionable government activities—the actual operation of economic enterprises (the Post Office, air traffic control, toll roads, TVA), all of which would be more efficiently operated as private firms, and a variety of unjustifiable subsidies, such as the provision of medical insurance to affluent old people, or the deductibility from federal income tax of interest on home mortgages. Government-run businesses and most government subsidies displace more efficient private activity, but regulation is essential and cannot be outsourced. Not that there isn't excessive regulation; but some—notably of financial markets—is indispensable.
Another potential confusion is between comprehensive economic regulation of specific industries and the regulation of safety and health and of workplace discrimination, cutting across industries. Public utility and common carrier regulation, illustrated by the regulation of telephone companies and railroads before the deregulation movement, was notably inefficient, tending to protect not consumers but instead sellers, by shoring up the sellers' cartels or monopoly. Banking regulation was of that character before the deregulation movement—limitations on the grant of banking charters, on branch banking, and on the payment of interest on demand deposits were examples of regulatory policies that reduced competition in banking. This was not an entirely undesirable effect because the more competitive banking is, the riskier it is—and the risks are macroeconomic (in contrast, if the airline industry, say, went bankrupt, the consequences for the rest of the economy would be trivial). But the traditional regulation of banking was too restrictive and was rightly dismantled—only the deregulation of banking (and related financial institutions) went too far.
The deregulation movement that focused on comprehensive regulation of specific industries coincided with a movement in the opposite direction—toward greater regulation—with regard to safety, health, pollution, and discrimination. Many of these regulations have no economic justification; they are paternalistic, as in the case of seatbelt laws—or if justified, are justified only because of the existence of other unjustified government interventions, such as subsidies for the medical expenses of people injured because they don't fasten their seatbelts.
But the fact that there is a great deal of unsound or questionable regulation is not a good argument for leaving all economic activity to the Darwinian processes of the market. Competition forces businesses to ignore external costs and benefits (that is, costs and benefits not borne by the creator of them). If either sort of externality is great enough, there is a strong case for regulation, provided the benefits of regulation can be shown to be highly likely to exceed the costs.
Market Failure Compared to Government Failure -Becker
When an industry in the private sector is not performing efficiently or effectively, there is said to be "market failure". The recommendation by economists and others typically is then for government actions to combat such failure, such as taxes to help reduce pollution. The diagnosis of market failure may be accurate, but the call for government involvement may be naïve and inappropriate.
The reason is that actual governments do not necessarily do what economists and others want them to do because there is "government failure" as well as market failure. Before recommending government actions to correct market failures, one should consider whether actual government policies would worsen rather than improve private sector outcomes. Since many factors often make for considerable government failure, considering such failure is crucial and not just a theoretical fine point.
Consider, for example, that consumers are sometimes ignorant of the qualities and other aspects of the products they buy. However, before advocating various forms of government protection of consumers, we should recognize that voters are far more ignorant of political candidates then consumers are of what they buy. The reason is that consumers directly suffer if they make bad choices out of ignorance, while individual voters have negligible influence over political outcomes. Hence voters have little incentive to be informed about different candidates and their positions, and the consequences of the mistakes they make are largely borne by others.
Monopolies do arise in the private sector, as when Microsoft had monopoly power over personal computer operating systems, when IBM still earlier had monopoly power over computers, or when manufacturers form cartels to raise their prices by restricting production. Yet, monopoly also occurs in the political sector, and it is far more pervasive there. An industry that contains only two firms is considered a duopoly that is presumed to raise prices above competitive levels, but the political process is dominated in democratic countries by duopolies, such as the Democratic and Republican parties. In addition, when government companies receive monopoly positions, such as the US Postal Service or national oil companies in many countries, they generally succeed in either keeping out or greatly delaying the entrance of private competitors. By contrast, private monopolistic positions are usually temporary, as seen in the eroding over time of IBM's and Microsoft's dominant positions in the computer industry.
Government actions sometimes not only fail to overcome market failure but rather worsen the failure. Fannie Mae and Freddie Mac were formed as quasi-governmental institutions to help encourage mortgages in the residential housing market because of a belief that the private sector was not providing enough mortgages, especially to lower income families. Yet, as documented in detail in Reckless Endangerment by Gretchen Morgenson and Joshua Rosner, these two companies used their privileged positions to take excessive risks, and to insure large numbers of mortgage loans that should never have been made.
European regulators have attacked Microsoft, Google, General Electric, Intel, and other (mainly American) companies because of various alleged anti-competitive policies. In these cases, and in many antitrust cases brought by American regulators, such as the recent objection to the merger of AT&T and T-Mobile, the motivation seems to be to protect the competitors of these companies or to protect jobs rather than to improve outcomes to consumers.
Many countries subsidize various alternative forms of energy, such as wind, solar, biofuels, and electric batteries, because of the substantial pollution from using coal, oil, and other fossil fuels. Often, however, the choices of what to heavily subsidize are made on political rather than economic criteria. For example, for years hydrogen cars were politically the most promising substitute for gasoline driven cars; then hydrogen fell out of favor and electric cars became the political darlings. Since governments have seldom succeeded in picking technological winners, I suspect they will be wrong again in these attempts to steer the development of cost-effective alternatives to the internal combustion gasoline engine. Another example is the scandal about the heavy American government financial support to the solar panel company Solyndra that recently failed.
How does one approach policy once it is recognized that government failure is substantial, and often much worse than market failure? As a general rule I believe the presumption should be in favor of government actions only when market failures are quite large and persistent. So clearly governments should have the dominant role in the military and police areas, in the judiciary, in protecting against massive pollution, and in providing a safety net for its least fortunate members (private charities are important but do not do enough). On the other hand, when market failures are relatively small and likely to be temporary, as in monopoly situations or in exploiting consumer ignorance, government involvement should be minimal, as in minimalist anti-trust policies, and in allowing consumers generally to make their own decisions.
The intermediate cases are the most difficult: when market failures may be significant, and yet government alternatives are not attractive. This may be decided on a case-by-case basis, but I believe the usual rule should then be to let the market operate. This belief is based on the conclusion that, on the whole, government failure is far more pervasive, damaging, and less self-correcting, than is market failure. Others may reach different conclusions, but these are the problems that a relevant welfare analysis should focus on. Simply concluding that in particular instances markets are not working perfectly is a misleading and incorrect basis for supporting active and sizable government involvement.