Regulation
and Economic Growth
Herbert Grubel
Professor of Economics (Emeritus)
Simon Fraser University
Draft of July 1, 2016
Abstract
At the initial stages of economic development
regulations increase productivity and create a positive relationship between
the stock of regulation and economic growth but this relationship is subject to
diminishing and ultimately negative returns At later stages of development regulations are
increasingly used to satisfy income-elastic demand for government-supplied social
benefits and public goods, which reduce growth in the market sector but
increase the growth of income defined to include these ephemeral benefits. This
model describes conditions existing in the advanced economies of the world.
The
study also discusses the difficulties inherent in the estimation of costs and
benefits of regulation, especially in the case of social benefits and public
goods. Public choice theory suggests that modern economies are over-regulated.
The
study implies that traditional policies to stimulate growth are bound to fail
because of the public’s demand for social benefits and public goods. More
political oversight and transparency is recommended to deal with public choice
issues around regulations.
JEL Classification: D61, D62, D63, D73, H10,
H21, H41, L51, O40
The secular decline in rates of economic
growth in the world’s developed countries since the 1970s has been analyzed
widely, especially after the unusually poor record of growth after the Great
Recession of 2008. Economists have suggested a number of causes such as income
inequality, fiscal austerity, excessive levels of private debt and the absence of
scientific and technological breakthroughs.
One cause mentioned by all studies of the slow
growth phenomenon mention excessive government regulations. This paper focuses
on this relationship because of its importance. According to Gordon and Katz (2015) “...regulatory excess
increasingly inhibits economic growth. Unless constrained, the regulatory state
will overwhelm America’s entrepreneurial spirit...”[1]
In the first part of this paper I argue that for
many years in the past the relationship between regulations and economic growth
has been positive because regulations during that time tended to enhance
economic efficiency and productivity. However, in recent years regulations have
reduced growth rates because high average incomes resulted in a public demand
for regulations designed to produce social benefits and public goods, which decrease
economic growth measured in conventional national income accounts but increases
growth in non-measured dimensions of income
Part two considers the implications of the
change in the dominant goal of regulations from increasing efficiency to
creating social benefits and public goods. My model suggests that a marginal
increase in regulations can reduce conventionally measured growth rates but
increases rates of growth in income that includes social benefits and public
goods. This result is consistent with recently observed conditions in developed
countries.
The third part of this study discusses the
problems associated with the valuation of the costs and benefits of regulations,
which make unreliable all estimates but especially those providing social
benefits and public goods.
In part four, public choice theory and the
theory of bureaucracy are used to establish a strong presumption that the
incentives facing politicians and regulators lead to excessive amount of regulation
for the creation of social and public goods.
The concluding section considers the policy
implications of the preceding analysis and reviews policies that can and have
been used to curtail over-regulation.
The Optimum Level of Regulation
Consider a primitive economy that initially
is without a government in which economic growth is determined solely by
increases in population and the accumulation of capital in a fixed amount per
person. The absence of regulations protecting property rights prevents productivity-enhancing
innovations and increases in the stock of capital.
In Figure 1 the annual rate of economic
growth and the existing level of regulation are measured along the vertical and
horizontal axes, respectively. Under the preceding assumptions, the level of
regulation is zero and the annual rate of growth is assumed to be equal to OA.
Figure 1
Now assume that the residents of this economy
form a government and give it the authority to establish regulations to protect
property, create a police force to enforce them and collect taxes to cover the
costs of these policies. The protection of property encourages the accumulation
of capital and entrepreneurs to develop new productivity-enhancing technologies
and products. The resultant gains in productivity and output are greater than
the government and private costs of the regulation. This development is reflected
in the upward slope of the curve in Figures 1.
The growth in the size and complexity of the
economy following the protection of property is assumed to lead to the adoption
of ever more comprehensive productivity-enhancing regulations like those that govern
road traffic, standardize units of measurement and money, protect intellectual
property and increase education.
This positive relationship is assumed to
continue until diminishing returns to regulations set in. This development is
reflected in the levelling of the curve in Figure 1 in line with the higher
levels of regulation. At the level of OQ regulations the marginal costs and
benefits are equal and the optimum level of regulation is reached at the OB
rate of economic growth. Further
regulations generate costs greater than benefits and lower the growth rate
until it becomes zero at OR regulations.[2]
The growth depressing regulations are
especially severe when they reduce the innovative activities of entrepreneurs[3] and the
high costs of compliance on established private businesses makes some non-profitable
so that the labour and capital they used are redeployed in uses where they are
less productive.[4]
New Goals of Regulation
The negative effect of regulations on
economic growth is increased as average incomes rise and the people demand social
benefits that are not sold in private markets such as protection from the loss
of income due to ill health, unemployment, retirement and disabling accidents.
The people also demand that the government supply public goods such as human
rights for minorities, a more equal distribution of income and wealth, the
creation of desirable industries and the protection of consumers, the
environment and culture.
The value of these social benefits and public
goods are not included in national income estimates based on existing
accounting conventions but their production requires labour, capital and
generates losses in the efficiency of the private sector and thus reduces
conventionally measured economic growth.
A Different Optimum Level
In Figure 2 the horizontal axis measures the
stock of regulations. The vertical axis on the left side represents growth
rates for GDP measured according to present accounting conventions. The right
vertical axis measures growth rates for GDP including the value of the currently
non-measured benefits from the consumption of social and public goods.
In Figure 2 the solid line AR replicates that
shown in Figure 1. The broken line AT shows
the relationship between regulation and the new index of national income that
includes social and public goods as well as the traditionally measured income.
The two lines coincide until regulations are at OX. Thereafter, the two curves diverge and the
curve reflecting additional value of social and collective benefits is a broken
line.
I hypothesize that conditions existing in the
real world are described in Figure 2 when the level of regulations is at OS,
the growth rate of conventionally measured income is OD and the expanded
measure of income is at SC. These conditions are the result of the public’s
demand for income-elastic social and public goods, which use resources that
cause conventionally measured growth to be low.
The preceding analysis has important policy
implications. The low levels of economic growth of recent times that has
attracted so much attention cannot be cured by the traditional policies
stimulating investment and innovation. The slow rates of growth are the cost of
creating social and public goods, which absorbs the capital and innovative
efforts that otherwise would be used to increase the output of conventional
goods and services.
Unfortunately, the data needed to test the
validity of the model presented here do not exist. The costs of regulations
bringing both efficiency-enhancing and social and public goods have been
measured but the measurement of the benefits is elusive, especially those
involving the ephemeral nature of the social and public goods. However, for
reasons discussed in the next section, there exists a strong presumption that
these benefits are over-estimated and have led to inefficiently high levels of
consumption and regulation.
Figure 2
Estimating the Costs and Benefits of Regulations
In a number of studies the level and cost of
regulation is assumed to be proportional to the number of regulations in
effect. A variant of this approach counts the number of pages required to list
existing regulations in the document published by regulatory agencies.
The indices created by this basic methodology
suffer from the fact that regulations are very heterogeneous. For example, a regulation
restricting the use of a small area of farm land to protect an endangered
species costing $1 million annually counts as much as another one-page regulation
prohibiting the use of an herbicide affecting millions of farmers and costing
billions.
For this reason, the proper measurement of
the level of regulation should weight each by the costs it causes. These costs
consist of spending on the creation, enforcement and policing of regulations by
the government agencies plus the cost of compliance by regulated enterprises
and the economic cost resulting from the shift of resources out of their most
efficient use in the private sector.
The basic problem with this methodology is that
estimates of the costs of regulation encounter a well-known accounting problem.
Government agencies and private firms have large fixed costs and no objective
and universally acceptable principle exist for allocating proportions of this fixed
cost to specific regulations. This fact is important because it can lead to systematic
biases in the estimates of costs made by civil servants with ideologically
driven policy agendas.
Additional problems with the measurement of
the costs of regulation arise from the fact that some require the use of
sophisticated mathematical models that employ many assumptions about future
developments, which makes the estimates highly uncertain and subject to the
introduction of ideologically driven bias.
Economic costs resulting from the effect of
regulations on entrepreneurial activities and innovation are especially
difficult to measure since statistical surveys cannot obtain information about
the extent to which the regulations prevent entrepreneurs from bringing their
innovations to market or are completely discouraged from even starting their
work.
In spite of these problems with estimates of
the costs of regulation, a number of studies have calculated them. A recent study is by Clyde Crews (2015) who
found that for the US economy in 2014 the costs came to $1.88 trillion, or
about one half of federal government spending that year.
The Benefits of Regulation
The benefits of regulations are even more
difficult to quantify than the costs. For example, the regulation forcing the
installment and use of seatbelts in automobiles reduces injuries and deaths.
Savings in medical costs and the preservation of incomes of accident victims from
the use of seatbelts can be estimated, but what is the value of the avoided
pain and suffering?
Estimates are even more difficult in the case
of regulations producing social benefits. For example, what is the value to individuals
and society from knowing that their incomes are made more secure through the
public provision of health care and pensions?
The value of the benefits from regulations
producing collective goods is also impossible to estimate. For example, the
value to individuals and society resulting from regulations that reduce the discrimination
minorities and women encounter in economic and cultural activities is ephemeral
and cannot be measured. Another example is the value of the benefits from the
prevention of global warming. The forecasts of temperatures with and without
regulations are very uncertain as are their effects on extreme weather
incidents, crop yields and property damage.
It is highly unlikely that the problems of
producing reliable estimates of costs and benefits of regulations will ever be
overcome and the existing, imperfect measure will be used to provide at least
some useful information for analysts. However, the next section presents
theoretical considerations, which imply that the cost/benefit calculations made
by the regulatory agencies are biased in favor coming up with net benefits for
specific projects, which lead to the adoption of regulations that objective
evaluations would have prevented.
Biased Estimates of Costs and Benefits of Regulation
In democracies politicians are expected to
pass legislation only if it improves the well-being of society after due
consideration of the resulting costs. Civil servants in ministries and regulatory
agencies design regulations needed to put into effect such legislation, with
each regulation implemented only if its benefits exceed its costs.[5]
It has long been assumed that politicians and
civil servants make these cost/benefit calculations motivated only by the
desire to maximize the well-being of society. However, this assumption has been
challenged not so long ago by the theory of public choice developed by James
Buchanan and Gordon Tullock (1962) and the theory of bureaucracy by William
Niskanen (1971) (1975). Buchanan and Tullock argued that politicians are
motivated not just to improve the welfare of society but also by the desire to
be elected or to stay in office and government. Niskanen argued that the
motives of civil servants in the evaluation of regulations are influenced by their
desire to enjoy a successful career, high pay, good reputation and important influence
on events.
The
work of politicians involves a difficult calculus. On the one hand, they are
constantly besieged by groups of citizens and organized businesses with demands
for specific laws and regulations allegedly improving the well-being of all
citizens. On the other hand, they face the risk that the vast majority of
citizens who have to pay the cost of the benefits to the interest may oppose
the legislation. While the interest group guarantees the politicians votes and
financial support, if only a relatively small proportion of the rest of the
population withdraws its electoral support, the politicians’ re-election
chances can be doomed.[6]
Buchanan’s analysis suggests that individual
voters either do not know about the cost such legislation imposes on them or
they believe the cost to be so low that it is rational for them to ignore the
issue. As a result, much legislation serving mainly the interest of relatively
few citizens is passed even if politicians know that a rational cost/benefit
analysis shows it to reduce overall public well-being and a properly informed
electorate would vote against it.[7]
Civil servants in charge of designing,
implementing and policing regulations are also subject incentives that bias
their decisions. On the one hand, if they approve the marketing of a new
product that turns out not to be effective or safe, their careers and incomes
are threatened. They therefore delay their approval until research indicates
that it is effective and safe at a probability much higher than it would they
did not face these personal costs. On the other hand, the delay of their
approval imposes costs on the owners of the new product and on consumers who would
have benefited from buying it. These costs do not enter into the civil
servants’ personal calculus determining the conditions for licensing a product.
This problem is especially serious in the case of pharmaceuticals, where the
delay in the approval of new medications can cause serious, unnecessary
suffering by patients with painful and life-threatening diseases.
Finally, it is worth mentioning that
over-regulation also takes place because civil servants and politicians are
entering these occupations because they believe in the need for governments to
deal with the shortcomings of free markets, which they see in producing great
inequalities in income and enabling large corporations to exploit consumers and
spoil the environment. Given the uncertainty surrounding estimates of costs and
benefits of regulations affecting these alleged problems with free markets,
their views are likely to under-estimate costs and over-estimate benefits.
Policy Implications
The main policy implication of this study is based
on the conceptual framework, which distinguishes between regulations that
increase efficiency and economic growth and regulations that bring social
benefits and public goods. The shift from the first to the second type of
regulation in recent times caused economic growth measured traditionally to
slow but has increased growth of income including social benefits and public
goods.
This result implies that the existing low
rates of economic growth cannot be improved by policies that stimulate capital
formation, education and innovation as the increase in these factors of
production is likely to be used to meet the public’s demand for more social and
collective goods.
It is an open question whether the present
mix of investment in the two types of output meets the true preferences of the
public because the trade-off between them is not known to the public and
politicians and has not been a major issue in election campaigns. The need for
such discussion is especially important to counter the populist notion
presented by the leading candidates in the 2016 presidential election who argue
that economic growth can be returned only by protectionism, isolationism and
income redistribution, economic planning and other illiberal policies. These
policies will be costly and fail to achieve the desired objective.
The second half of this study implies that
the public discussion of the proper mix of regulations aimed at the two types
of benefits should take account of the strong presumption that the motives of politicians
and civil servants for private gains has resulted in an excessive production of
social benefits and public goods.
A broad public discourse on the optimal mix
of regulations is important but likely to take a long time to improve political
decisions and regulatory practices. In the meantime, governments can adopt
policies that ensure that the analysis of costs and benefits is correct and
protected from the biases stemming from the selfish motives of politicians and
civil servants.
Leading Washington think tanks have done much
research, published many studies on regulatory policies and presented proposals
for reforms of the regulatory system: The CATO Institute has for more than 25
years been publishing in its periodical Regulation
Magazine papers on regulation reform by distinguished researchers; The
Competitive Enterprise Institute has been publishing periodic updates focusing on
the cost of regulation, the latest of which is by Crews (2016)); and The Heritage
Foundation published a volume (Gordon and Katz (2015)) containing
recommendations for policy reforms written by a number of experts.
Almost all of the published studies on
regulation criticize specific policies on the grounds that the costs exceed the
benefits. These studies are valuable and can serve as useful guides for changes
in specific policies if and when the people of the United States elect a
government willing to consider such changes.
More important are the recommendations found
in these studies for fundamental changes in the way the costs and benefits of past
and future regulations are determined: The primary responsibility for this task
should be returned to the elected members of Congress and away from the
unelected the civil servants operating regulatory agencies. To protect the
process from politicians attempting to use it for personal gains, the work of
Congress should be made transparent and accessible to the media and public
scrutiny.
Under these proposed reforms elected officials
would face mountains of additional work organizing and attending relevant committee
hearings and voting on proposed regulations. Their workload could be eased by
restricting the Congressional evaluation of regulations to only those with a
likely cost of a certain level, such as $500 million.
Another policy change would be for Congress
to invite public petitions for review of specific regulations regardless of the
costs estimated by the regulatory agencies. This process should be triggered
only if the petitioners exceeded a specific number set high enough to deter
frivolous use of the instrument. Such reviews would create an important
obstacle to the existence of regulations for which costs exceed benefits and
would encourage regulatory agencies to be more diligent in their own research
on the costs and benefits they design and consider for adoption.
Strong and effective Congressional oversight
of regulations is not much different from that required of military, social and
other spending programs. The results of these efforts are unlikely ever to be
perfect, but they would almost certainly increase the chances that regulations
are efficient and respond to the demand of the general population.
Another approach to reducing the level of
regulation in the economy involves using market signals and competition. Chris Edwards
(2016) recommends privatization of publicly owned enterprises so that market
forces reveal the damage done by regulations and lead to their abandonment
rather than rescue by further regulations. Privatization efforts have met with
some success but run into powerful opposition from interest groups benefiting
from public ownership.
Grubel (1983) suggests the creation of free
market zones in which regulations are at a bare minimum and whose commercial
success encourages regulated businesses to demand equal treatment. The most
successful experiment in this spirit has been the creation of special economic
zones in China. The success of the first zones has inspired the creation of
many others, led to the reduction in regulations throughout the country and provided
the conditions for China’s economic growth. Free market zones have also been
created successfully in some developing countries, but few can be found in
advanced market economies where political and interest group opposition is
pervasive.[8]
References:
Becker, Gary (1976), “Toward a More General Theory of Regulation”, The
Journal of Law & Economics, Vol.
19, No. 2, pp. 245-248 found at https://www.jstor.org/stable/725165?seq=1#page_scan_tab_contents
Buchanan, James M. and Gordon Tullock (1962), The
Calculus of Consent: Logical Foundations of Constitutional Democracy, Ann
Arbor: University of Michigan Press, reprint found in
June 2016 at http://www.econlib.org/library/Buchanan/buchCv3Cover.html
Chao, Johnny and Herbert Grubel (1998),
“Optimal Levels of Spending and Taxation Canada”, pp. 53-68 in Grubel editor, How to Use the Fiscal Surplus: What is the
Optimal Size of Government, Vancouver: The Fraser Institute found at https://www.fraserinstitute.org/sites/default/files/HowtoUseFiscalSurplustblContents.pdf
Crews, Clyde Wayne Jr. (2016), Ten Thousand Commandments: An Annual Snapshot
of the Federal Regulatory State 2015 Edition, Washington: Competitive
Enterprise Institute, accessed June 13, 2015 at https://cei.org/sites/default/files/10%2C000%20Commandments%202015%20-%2005-12-2015.pdf
Edwards,
Chris (2016), Privatizing the U.S. Postal
Service, Tax and Budget Bulleting No. 75, Washington: CATO
Institute found at http://www.cato.org/publications/tax-budget-bulletin/privatizing-us-postal-service
Gordon, Robert and Diane Katz, Editors (2015),
Environmental Policy Guide: 167
Recommendations for Environmental Policy Reform. Washington, DC: The
Heritage Foundation, accessed June 2016 at http://thf_media.s3.amazonaws.com/2015/pdf/EnvironmentalPolicyGuide.pdf
Grubel, Herbert
(1983), Free Market Zones: Deregulating Canadian Enterprise, Vancouver: The
Fraser Institute found at https://www.fraserinstitute.org/sites/default/files/free-market-zones.pdf
-------------------- (2016),
“Canada’s New Immigration Policies 2016: Good Politics, Bad for the Country”, Optimum Online: Journal of Public Sector Management, 46, 2, summer; found at http://www.journals4free.com/link.jsp?l=29062666
Laffer, Arthur (2004), The Laffer Curve: Past, Present, and Future, Backgrounder #1765 on Taxes,
Washington: Heritage Foundation, found at http://www.heritage.org/research/reports/2004/06/the-laffer-curve-past-present-and-future
Niskanen, William
A. (1971), Bureaucracy and
Representative Government, Aldine, Atherton
--------------------------- (1975) “Bureaucrats and Politicians”, The
Journal of Law & Economics, Vol.
18, No. 3
Regulation Magazine published quarterly by the
CATO Institute, found at http://www.cato.org/regulation/about
Scully,
Gerald (1989), “The Size of the State, Economic Growth and the Efficient
Utilization National Resources”, Public
Choice 63, pp. 139-164
[1]
The quote is from Gordon and Katz (2015) page 31.
[2]
The shape of the inverted U with the right end of the curve at zero growth
rates used here is the same as that found in the study by Laffer (2004), which
has tax revenue on the vertical axis and the level of taxation on the
horizontal axis and that by Scully (1989), which has economic growth and
government spending on the two axis. The study by Chao and Grubel (1998)
contains an empirical estimate of the Scully curve for Canada.
[3] The effects of regulation on entrepreneurial innovations, productivity
and economic growth are nearly impossible to measure. The legendary
entrepreneurs that created computers and other successful products and services
in their garage came to the attention of statistical agencies only after they
marketed their products successfully, employed workers and capital and paid
taxes. There are no records of the entrepreneurs who stopped their work because
the regulatory requirements and delays caused them to run out of money or of
the entrepreneurs who were discouraged from pursuing their dreams because they
knew of the regulatory costs they would face.
[4]
The economic costs of regulation caused by the closing of businesses are also
nearly impossible to measure because such closures take place all the time for
other reasons and in practice it is very difficult to identify the role played
by the cost of regulation and other factors affecting profitability.
[5]
The study by Gordon and Katz (2015) indicates how policies set by law leads to
regulations designed and executed by regulatory agencies. For example the
“Clean Air Act” and “Clean Water Act” authorize the Environmental Protection
Agency (EPA) of the United States to design, enforce and policy regulations
needed to achieve the objectives of these acts.
The Bureau of Land Management, the U.S. Forest Service, the U.S. Fish
and Wildlife Service, and the National Park Service pass regulations to achieve
the objectives of the “Endangered Species Act” and some provisions of the
“Clean Water Act”.
[6]
Gary Becker (1976) suggests that the political system leads opposition
politicians to make the public aware of the costs of special interest
legislation passed by the government. This effort by opposition politicians
represents a restraint on the ability of politicians to buy votes and on the
size of the costs their legislation imposes on the population.
[7] Current
immigration policy in Canada illustrates this proposition. According to the
analysis found in Grubel (2016) the government headed by Justin Trudeau elected
in 2015 increased the annual admission of immigrants from 270,000 to 300,000.
Canada’s large immigrant communities benefit from this new policy as it
strengthens their political and economic influence in the country and allows
them to grow their cultural traditions and institutions. This legislation
brings the Trudeau government many votes from the immigrant communities, which
in some electoral districts can decide the outcome of federal elections.
The government does not risk a significant loss of votes from the rest
of Canadians because they do not know that the legislation imposes serious
costs on them or because they have been persuaded that immigrants increase
economic growth, prevent labour shortages and create a favorable demographic
balance need to finance social insurance program.
In fact, however, these benefits are illusory while recent immigrants
impose a heavy fiscal burden of $30 billion annually on taxpayers, mainly
because recent immigrants pay taxes much below the national average but consume
government services equal to the average. Immigrants also aggravate the problem
of income inequality as they raise returns on capital and lower the incomes of
labour. In addition they increase the cost of housing and the levels of
pollution, traffic congestion and shortage of medical services.
[8]
When first I published some studies on this subject and it attracted interest
from the media and some jurisdictions in Canada, a powerful mandarin in the
federal government wrote to me that he and his government opposed the creation
of free market zones because eventually they would spread throughout the
country.
.jpg)