February 20, 1991

February 20, 1991 | Lecture on Political Thought

Constitutional Economics: The Framework for Economic Growth and Social Progress

(Archived document, may contain errors)

Constitutional Economics: The Framework for Economic Growth and Social Progress

By Gerald W. Scully

S ome thirty years ago, Robert Heilbroner, the doyen of the leftist New York intelligentsia, wrote that the matter was settled that socialism would bring a standard of living and social justice to mankind greater than is possible under a system of capitali s m, free markets, and individual freedom.' Recently, he said that the evidence from the 75 year struggle between the systems was that capitalism has won.2 While for Heilbroner this is an epitaph of a wasted intellectual life, the debate and the destiny of mankind under these two competing systems hardly is over.

Conservatives have seen in the de-Stalinization of the Soviet Union and in the agrarian economic reforms of the People's Republic of China the promise of a sharp move awayfrom statism. Nevertheless, these reforms are from the top down: mandated by the Communist Party. None of the reforms call for private property, freedom of contract, free markets, and the rule of law. The reforms brought about by the spontaneous uprisings of the peoples of Eastern E urope are more promising. At this point, the peoples' demand is for a say in their political system. The institutions of private property, free markets, and the rule of law that are crucial to the functioning of a free society need to be built from scratc h . These people seem united in their opposition to the socialist model of man and society. It is not clear that they are prepared to move fully to a model of man as independent and responsible. Ile short- run costs of adjustment from the move from a contro lled to a freer economy are high and politically dangerous. These costs of adjustment will test the patience of the citizens of Eastern Europe and offer opportunities to politicians for the resurgence of the idea of government intervention.

I Robert L. Heilbroner, The Future as History (New York: Harper, 1960). 2 Robert L. Heilbroner, "The Triumph of Capitalism," The New Yorker 64 (January 23, 1989): 98- 109.

D r. Gerald W. Scully is a Bradley Resident Scholar at The Heritage Foundation and Professor of Economics at the University of Texas, Dallas. He spoke at The Heritage Foundation on December 14, 1990. ISSN 0272-1155. 1991 by The Heritage Foundation.

And, all of the world is not Europe and North America. Much of mankind, on the order of three-quarters, lives under dictatorship, with state control of the economy. The political revolution of Eastern Europe has not travelled to Africa, Asia, or much of Latin America.

Predictions on the Pattern of Growth in the Neoclassical Model. For the past thirty years , the main tool of analysis in the theory of 3economic growth has been the neoclassical model developed by Robert Solow. In that model, the per capita growth rate is linked to the rate of accumulation of the capital stock. For the simplest set of assumpti o ns in the model, economies evolve from a low capital-labor ratio (low per capita income) to a high capital-labor ratio (high per capita income) state determined by the population growth rate, the savings rate, and the returns to capital. As with the assum p tion in classical economic theory, the returns to capital fall as capital accumulates. Eventually a steady state capital-labor ratio prevails, and the real rate of per capita income grows at the rate of population growth. While the neoclassical growth par a digm of a steady state growth rate (barring further innovation or resource discovery) is dismal, it is not as dismal as in the classical theory (Smith, Malthus, and Ricardo) where the steady state in capitalist economies is a level of per capita income at the subsistence level.

Table 1. Economic Growth Rates and Measures of Freedom

Statistic (1) (2) (3) (4) (5) (6)

Economic Growth Mean 1.3 3.2 3.4 2.1 1.7 2.3 Standard Deviation 1.9 3.4 0.9 1.1 1.5 1.1 Political Liber Mean 5.7 4.3 1.2 2.8 3.2 2.2 Stand ard Deviation 1.4 1.6 0.5 2.2 1.8 1.6 Civil LibgM Mean 5.6 4.7 1.7 3.0 3.5 2.4 Standard Deviation 1.1 1.7 1.1 1.5 1.4 2.1 Rank of Economic Liberjy Mean 91.5 78.6 29.0 38.8 67.8 31.1 Standard Deviation 29.7 31.4 20.7 27.6 24.6 33.6

Legend: (1) Africa, (2) Asia, (3) Europe, (4) Central and North America, (5) South America, and (6) Oceania.

3 Robert M. Solow, "A Contribution to the Theory of Economic Growth," Quanerly Jounial of Economics 70 (1956): 65-94.


We have comparable data, now, on real national product for 130 countries of the world. Coverage is from 1950 to 1985, but for much of Africa the data begins in 1960. The average annual growth rates of per capita real gross domestic product are given by co n tinent in table 1. For all of the countries of the world the simple average growth rate is 2.3 percent. The range in economic growth is extreme. More than a quarter of the countries, mostly the poor ones, have real per capita growth rates of less than one percent. To put the sorry state of the poor countries of the world into perspective, at a growth rate of one percent it takes three generations to double real per capita income. An implication of the neoclassical growth model is that growth rates will con v erge across countries through time. If we compare the growth rate of the economy with its past value (say in 1950 or 1960) we ought to observe that the low per capita income countries grow at a faster rate than the high per capita income countries, becaus e of the difference in the marginal rates of return to capital. In fact, we. observe the opposite, contrary to the predictions of the neoclassical model. High per capita income countries are growing at a faster rate than low per capita income countries. Va r ious studies have shown that the apparent rate of return to capital in the less developed world is a multiple of the rate of return in the developed world. Here we have a paradox that neoclassical theory cannot explain. If the apparent rate of return to c a pital is so high in the less developed world, why'don't we see massive capital transfers in their direction? In fact, we see capital flight from the less developed world to the developed world. Therefore, something other than purely economic factors must b e at work. The Constitutional Setting and the Gains from Exchange. Behavior must be sanctioned and rights protected for economic progress to occur. The scope of sanctioned behavior and of rights are contained in the constitutional setting or institutional framework of society. Ile rules and institutions of a society may evolve spontaneously (e.g., common law) or may be created by government authority (e.g., statute law). These rules and institutions may be efficient or inefficient. Pareto efficient rules o r institutions are those that define a set of sanctioned activities that when undertaken enhance private wealth or utility. Pareto efficient rules have the attribute that all members of society benefit from their existence. As examples, the introduction of a commodity based money as a substitute for barter exchange facilitates the gains from exchange and reduces transaction costs; or the rise of an insurance and a stock market diversifies risk. Rules or institutions in which the total gains to society excee d the total losses are termed Hicks-Kaldor efficient. Since some gain at others' expense, such rules are redistributive. Absent costs of redistributing income, a Hicks-Kaldor efficient rule can be converted into a Pareto efficient rule. Examples might incl u de fiat money and bankruptcy laws. Finally, there are rules and institutions in which the total gains to society are less than the total losses. These rules are termed Hick-Kaldor inefficient and they sanction behavior that is termed rent-seeking. Example s include trade unions, farm price supports, protective tariffs and, to the extent that the fiscal institution is a redistributive mechanism perhaps, Congress. The constitutional setting serves three basic functions: protection of rights, provision of publ ic goods, and redistribution of income. These functions variously structured have


efficiency (wealth creating) and equity (income distribution) attributes. 'Me choice between efficiency and equity within a constitutional setting hinges on decisions ma de with respect to the division of powers between citizen and the state and on the type of legal system. Conventionally, the division of power (rights) between the ruler and the ruled rests on the constitutional contract (document) and on the standing of t he citizen relative to the state before the law. In general, law that arises from the custom of exchange and human intercourse (common law) fosters private wealth maximization and minimizes rent-seeking (income redistribution). The common law is a "common s " and as such is very costly to privatize for special interest rent-seeking. Statute law is made in a political market, subject to the political will of a sovereign majority. Statute law potentially is redistributive because under universal suffrage (one m an, one vote) and a skewed income distribution, income redistribution platforms draw a majority of the voters. Legislators seek or retain office by rewarding the coalitions that elect them. Hence, law by legislation is more likely redistributive than is p r ivate law. Furthermore, since one legislature cannot bind another but precedent (stare decisis) binds judges, statute law is more uncertain than judge-made law. Uncertainty weakens the incentives to accumulate capital or engage in long term commitments. H ence, over the long run, it reduces economic efficiency and growth.

In James Buchanan, The Limits of Liberty, a constitution of equal rights for all, in a two-person setting, emerges from the leap from the Archimedean point of Hobbesiain anarchy. In figure 1, A is the utility levels of person 1 (Ul) and person 2 (U2) in the natural state - the Hobbesian jungle of every man against every man. Utility is low, because what can be produced and consumed is severely constrained by the need to devote resources fo r defense and predation. By definition anarchy is the absence of mutually agreed and respected rights. The income distribution in anarchy strictly is a function of the relative natural distribution of physical strength, stealth, and guile of the conflictin g parties.

The cooperative constitutional contract (point F in the figure) is the Pareto-frontier reached through an agreement on equal freedoms. The constitutional contract of equal rights permits efficiency gains from trade (specialization, economies of scale, etc.) in the postconstitutional stage (point G in the figure). The income distribution in this postconstitutionaI stage is strictly determined by unequal personal endowments (talent, ambition, motivation, skill, luck, choice of one's parents, etc.) that result in unequal economic outcomes.

Between the regimes of anarchy and equal rights there are many "constitutional contracts" in which rights are distributed asymmetrically among the parties (e.g., a master-slave society, feudalism, socialism, etc.) . Nevertheless, agreement among unequals (in rights) is welfare-superior to the anarchy at A in the figure. In general, 66constitutional contracts" can be arrayed hierarchically from the most inequitable distribution of rights or opportunities to compete f or income streams to the most equal distribution of rights (i.e., B,D,F in the figure). Each "constitution" has an income distribution given partially by the rights distribution (opportunities to compete for income streams). Each "constitution" has a post constitutional stage of gains from trade. Hence, each "constitution" has an inherent level of economic efficiency.




I G I I I I I I ------ I I F I I E I I I I I I I I D I I I I - c ---- I I-- B_ A


Figure 1





A z w B

U- LL B w U- 0 x- A w D 0 z



0.0- MAXIMUM R 0"R0


Figure 2


Figure 2 describes the paths of economic efficiency and equity or the distribution of rights to compete for income streams of the hierarchy of "cons titutional contracts." The agreement at B is characterized by wide disparity of rights among the groups. Because rights are distributed unequally, incomes are distributed unequally. Gains from trade are limited. As such, economic ekiciency under this cons t itutional contract is low. Constitutional contracts with less inequality of rights (e.g., D) are associated with more equality of opportunity to compete for income streams and, hence, wider gains from trade in the postconstitutional stage. Hence, greater i ncome equality and economic efficiency are present in regime D than in regime B. In the constitution of equal liberty for all (F in the figure), inequality of rights to compete for income streams is zero and economic efficiency is at a maximum (a value of the index equal to one).

Evidence on the Effect of the Constitutional Setting on Economic Growth and Income Distribution. I have told an abbreviated- theoretical story about the relationship between the choice of the structure and distribution of rights and its consequence to economic efficiency, gains from trade, or economic growth and to income in equality. With suitable measures of the constitutional setting at hand we can estimate the effect of the constitutional setting on economic growth and on the income distribution.

Data on some of the institutional characteristics of countries of the world h ave been developed by Raymond D. Gastil of Freedom House. He has published annually, since 1973, country rankings of political liberty and civil liberty, and, for some years, measures of the type of economic system and the degree of economic freedom. Ile p olitical rights measure, scaled from 1 = free to 7 = not free basically is a measure of the degree of political competition and the right of the citizens to choose their leaders. The civil rights measure, similarly scaled, is a measure of the rule of law a nd the independence of the judiciary. The economic regime is a variable scaled from 1 to 9 that ranges from purely capitalist, free market systems to purely socialist, command economies. There is considerable skepticism about Gastil's political rights and economic rights measures, because they are majoritarian and biased in favor of human or positive rights. 4 Nevertheless, these measures are the most comprehensive that we have at hand as measures of the variation in individual rights around the world. A c r ude way of comparing the degree of liberty and economic growth is to examine simple comparisons of the average growth rate in real per capita domestic product over the period 1950 to 1985 with average values of political, civil, and economic freedom. This is done in table 1. Africa has the lowest rate of economic growth and the least amount of individual freedom. Europe has the highest level of freedom and the highest rate of economic growth. Asia has relatively little political freedom or civil rights but in some cases (Hong Kong, Taiwan, Japan) a high degree of economic freedom. Coupled with its outward trade orientation, economic freedom has led to a high rate of economic growth. A more empirically sophisticated method of examining the relationship betwe en economic growth and freedom is through the specification and estimation of a growth

4 Gerald W. Scully and Daniel Slottje, "Ranking Economic Liberty Across Countries," Public Choice, in press.


model that includes these measures of the institutional framework as arguments. This I have done in a published article.5

Table 2. Average Growth Rates of Per Capita Real Gross Domestic Product by Institutional Attribute

Institutional Growth Institutional Growth Difference Attribute Rate Attribute Rate in Growth

Politically Open 2.5 Politically Closed 1.4 1.1 Individual Rights 2.8 State Rights 1.2 1.5 Free Market 2.8 Command 1.1 1.7

A summary of the results is given in table 2. On average, politically open societies grew at a rate of 2.5 percent per ann um compared to 1.4 percent for politically closed societies. On average, societies that subscribe to the rule of law grew at a 2.8 percent rate compared to 1.2 percent in societies where state rights take precedence over individual rights. On average, soc i eties which subscribe to private property rights and a market allocation of resources grew at a 2.8 percent compared to a 1.1 percent rate in nations where private property rights are circumscribed and the state intervenes in resource allocation. Thus the institutional framework is a phenomenon of considerable magnitude to the standard of living in a society. Growth rates in societies which circumscribe or proscribe political, civil, and economic liberty are less than half of those societies in which indiv i dual rights are protected. I have said that the distribution of rights has as much, and, perhaps more, to do with the observed distribution of income across countries than does differences in economic outcomes, holding the rights structure constant. A sop histicated analysis is beyond the scope of this lecture, but those interested can read a published paper of mine on the subject.6

5 Gerald W.Scully, "The Institutional Framework and Economic Development," Jounial of Political Econonzy 96 (June, 1988): 652-662. 6 Gerald W. Scully, "Rights, Equity, and Economic Efficiency," Pablic Choice, in press.


Table 3. Income Shares by Institutional Attribute

Measure Q1 Q2 Q3 Q4 Q5

Political LibeM Most Free 5.7 10.7 16.0 22.9 44.6 Least Free 5.3 8.0 11.5 17.9 57.4 Difference 0.4 2.7 4.5 5.0 -12.8

Civil LibeM Most Free 5.9 11.2 16.3 23.0 43.7 Least Free 5.0 7.4 10.6 17.3 59.8 Difference 0.9 3.8 5.7 5.7 -16.1

Economic Liberty Most Free 5.5 10.8 16.0 22.8 45.0 Least Free 5.6 8.0 11.1 17.7 57.6 Difference -0.1 2.8 4.9 ..5.1 -12.6

I can make the point with table 3 that shows the relationship between the quintile distribution of income averaged for 70 countries of the world and the classification of the rights structure from m ost free to least free. The evidence strongly suggests that free societies have higher shares of income going to the second through the fourth quintiles (20th to 80th percentiles) and lower shares being received by the fifth quintile (80th to 100th percen t ile). Conversely, societies in which political, civil, and economic rights are restricted have lower shares among the income recipients in the 20th to 80th percentiles and higher income shares to those in the 80th to 100th percentile. The relative share o f the poorest in society (Q1) is invariant to the choice of the institutional framework.'

In politically open societies compared to politically closed regimes the share of income to the middle three quintiles is 49.6 percent compared to 37.4 percent. In na tions that obey the rule of law compared to regimes in which the rights of the state are above those of the individual the comparison is 50.5 percent versus 35.3 percent. In countries that have private property, market resource allocation, and a minimum i nterference of the state in the economy compared to command economies the comparison is 49.6 percent versus 36.8 percent. Equally revealing as a matter of equity is the status of the poor and the rich in free


and in statist nations. The income share o f the highest income group is much larger in nations that repress individual rights than in societies where rights are protected. This is so, because substantial groups of citizens have been excluded from the opportunity to compete for income streams. Ave r aging across the rights measures the share of income going to the highest income quintile is 58.3 percent among the least free nations and is 44.4 percent among the most free, a staggering difference of nearly 14 percentage points. Among the poorest membe r s of society choice of the rights regime does not have much of an impact on their share of income. Certainly, the case cannot be made that the lack of or the surrendering of freedom buys more equity, as the socialists and purveyors of statism in the inter ests of social justice would have us believe.

Conclusions. What wisdom is to be gained from this research? John Locke and Adam Smith, the intellectual giants of a socio-political and economic philosophy of free men and free markets, had it right and the Gh andis, Nehrus, Nassers,- Nkrumahs, and other nationalists, and the development economists, like Gunnar Myrdal, got it wrong. The classical liberals have much to say to the human condition at the end of the 20th century. All societies have persons of talen t , ability, and ambition who are capable of transforming their lives, their families, and, in the aggregate, their nations from a rude state to one of a high level of economic and human development. And, a necessary condition for these universal human char a cteristics to be unleashed is a constitutional setting that fosters and protects private property, the rule of law, and allows for- competition among political agents who aspire to govern. There are many talented and ambitious people in the less developed world. Consider the contributions of Indians, Chinese, Soviets, Eastern bloc residents, Latins, Africans, and others in the pure sciences, mathematics, engineering, literature, art, sports, and a wide array of human endeavor. 71at there has been a lack of economic progress in these countries speaks not of the people but of their ideology and institutions: a failure to structure a constitutional setting that leaves people free to go about their business of self-betterment, unmolested by the state.

1 0


About the Author