The unbearable lightness of regulatory costs.

• Author: Ackerman, Frank

Will unbearable regulatory costs ruin the United States economy? This specter haunts officials in Washington, just as fears of communism once did. Once again, the prevailing rhetoric suggests that an implacable enemy of free enterprise puts our prosperity at risk. Like anti-communism in its heyday, anti-command-and-control-ism serves to narrow debate, promoting the unregulated laissez-faire economy as the sole acceptable goal and standard for public policy. Fears of the purported costs of regulation have been used to justify a sweeping reorganization of regulatory practice, in which the Office of Management and Budget (OMB) is empowered to, and often enough does, reject regulations from other agencies on the basis of intricate, conjectural economic calculations.

This Article argues for a different perspective: what is remarkable about regulatory costs is not their heavy economic burden, but rather their lightness. Part I identifies two general reasons to doubt that there is a significant trade-off between prosperity and regulation: first, regulatory costs are frequently too small to matter; and second, even when the costs are larger, reducing them would not always improve economic outcomes.

The next three parts examine evidence on the size and impact of regulatory costs. Part II presents cost estimates for a particularly ambitious and demanding environmental regulation, REACH--the European Union's new chemicals policy. Part III discusses academic research on the "pollution haven" hypothesis, i.e. the assertion that firms move to developing countries in search of looser environmental regulations. Part IV reviews the literature on ex ante overestimation of regulatory costs, including the recent claims by OMB that costs are more often underestimated (and/or benefits overestimated) in advance.

Turning to the economic context, Part V explains why macroeconomic constraints may eliminate any anticipated economic gains from deregulation. Part VI introduces a further economic argument against welfare gains from deregulation, based on the surprising evidence that unemployment decreases mortality. Part VII briefly concludes.

1. TWO ARGUMENTS AGAINST THE TRADE-OFF

   In theory, it would be possible to spend so much on environmental protection that basic economic needs could not be met. At a sufficiently high level of regulatory expenditures, protecting nature and cleaning up the air and water could absorb enough of society's resources to compete with the provision of more fundamental goods, such as food and shelter. From this, it is a short leap to the conclusion that the clash between economy and environment actually is an urgent problem, requiring detailed analysis of regulations to prevent worsening the terms of the trade-off. But the latter statement only follows logically if environmental policy is in fact consuming substantial resources, which are transferable to other, more basic needs. That is, the assumed urgency of the trade-off rests on the implicit assumptions that the costs of environmental protection are both large and fungible. Either of these assumptions could fail in practice; the costs of environmental protection could be nonexistent, or too small to matter, or the reduction of regulatory costs might not produce the desired economic benefits.

   Environmental protection with little or no costs

   Costless environmental improvement is frequently assumed to be impossible by definition. The hidden premise underlying this form of the trade-off argument is that the market economy is already performing as well as possible; that is, it has reached a Pareto optimum. (1) From this perspective, any new expenditure on environmental protection necessarily represents a loss, because it diverts resources away from the things that consumers, in their wisdom, have chosen for themselves. (2)

   Reverence for market outcomes is at odds with the beliefs of many environmental practitioners who assume that environmental improvements can bring economic benefits as well. The rhetoric of joint economic and environmental progress includes such overused imagery as "win-win solutions," the "double [or triple] bottom line," and opportunities to pick the "low hanging fruit." (3) The ubiquity of these phrases underscores the extent to which environmental advocates find that the market is improvable--implying that it could not have already been at an optimum.

   In a more academic vein, the Porter hypothesis maintains that carefully crafted, moderately demanding regulations can improve economic competitiveness and success in the marketplace. (4) Likewise, studies of energy conservation and greenhouse gas reduction frequently find opportunities for energy savings at zero or negative net cost, as in the "no regrets" options for climate change mitigation. (5) The critique of these opportunities is not that they are undesirable; who could argue with free environmental improvements? Rather, economists have argued that, in their own overused metaphors, there are no free lunches, nor twenty dollar bills on the sidewalk. (6) If lunch is expensive and the sidewalk is bare, then the Porter hypothesis must be impossible, and there must be hidden costs associated with energy conservation.

   Without attempting a thorough review of this debate, it seems plausible that there are significant cases where essentially costless energy savings and other environmental improvements are possible. In such cases, the fears of regulatory cost burdens and concerns about trade-offs are presumably easy to resolve; there should be a broad consensus supporting the adoption of costless improvements.

   However, literally costless improvements are not the only ones to escape from the trade-off; economic constraints do not immediately become relevant to real decisions as soon as regulatory costs are greater than zero. Very small costs of regulation presumably have very small impacts on the economy. Regulations could easily have costs that are too small to matter--and Parts II and IV will suggest that this is the case in many important instances. The theoretical consensus that supports costless environmental improvement may vanish once costs become positive, however small; but practical concerns about economic impacts need not arise until costs become large in some meaningful sense.

   The question naturally arises: what counts as large? Here it is important to resist the illusion of superficially big numbers. Quantities in the billions, which are commonplace in federal programs and nationwide impact assessments, (7) are essentially impossible to understand in isolation; some standard of comparison is needed to bring them down to a comprehensible scale. (8) Amounts in the billions of dollars are inevitably thought of as part of a ratio: if X billion dollars is the numerator, what is the appropriate denominator? When none is specified, the default denominator tends to be the listener's personal finances--in which case one or a few billion dollars appear very large indeed.

   In contrast, a penny per-person, per-day sounds small. But, for the United States, with its population of about three hundred million, (9) a penny per-person, per-day and a total of one billion dollars per year are roughly the same. (10) Per capita impacts, as in this example, are sometimes appropriate, particularly when the costs of regulations are spread across the population as a whole. Comparison to the revenues of the affected industry is also a useful standard for evaluating regulatory impacts. For issues affecting the entire United States, the European Union, or even a large industry, a few billion dollars or euros per year is not a large number. This issue is important in the discussion in Part II.

   Environmental costs that cannot be traded for economic gains

   Even when environmental policies impose noticeable economic costs, it does not necessarily follow that these costs could be traded for greater private incomes and consumption, or for the benefits that are thought to accompany higher incomes. There are two strands to this unfamiliar argument, presented in Parts V and VI below, and briefly anticipated here.

   First, deregulation might not produce increased economic growth. If a regulation or other environmental policy has measurable economic costs, it consumes resources such as labor and capital that could have been used elsewhere in the economy. The policy, then, can only be "traded" for whatever those resources could have produced elsewhere--in economic terms, the opportunity cost of those resources.

   During a recession, labor and capital are typically less than fully employed. Supplying more resources that are already in surplus may not produce anything more; the short run opportunity cost of additional resources could be zero. On the other hand, during expansions such as the late 1990s, the Federal Reserve (the "Fed") carefully controlled the level of employment and rate of growth; making more resources available for increased growth might just lead the Fed to step harder on the brakes in order to maintain the unchanged target pace of expansion. (11) Again, the short-run opportunity cost of additional resources could be zero.

   Second, economic growth may not produce the expected or desired benefits. An increasingly common method of analysis converts regulatory costs into health and mortality impacts, based on correlations between income and health. (12) In the extreme, regulatory costs that are thought to lower market incomes have been labeled "statistical murder," because richer people live longer. (13)

   This line of argument is flawed in several respects. Perhaps the most dramatic response to the "statistical murder" story is the epidemiological evidence that mortality decreases in recessions. If deregulation leads to economic growth, which boosts employment, the expected result is paradoxically not a reduction in mortality.

   In the long run, the availability of resources such as labor and capital...