COMPLEXITY, CHANGE AND ECONOMIC ANALYSIS

EMPIRICAL ESTIMATES OF BIASES RESULTING FROM OMITTING SOCIO-POLITICAL FACTORS IN ECONOMIC MODELS

THREE EMPIRICAL ESTIMATES OF THE SIMULTANEITY BIAS AND SPECIFICATION ERROR RESULTING FROM THE OMISSION OF POLITICAL AND SOCIAL FACTORS IN ECONOMETRIC MODELS

ABSTRACT

A complex interdependent world requires different modeling strategies than are conventionally pursued by economists. Because interactions among and between economic, political, social, and psychological variables are so significant; conventional utility maximization theory, OLS regressions, or simple two-equation systems inherently offer severely misleading answers with respect to questions regarding economic choice, economic development, and optimal economic policy. The biases can be severe enough to alter estimated directions of change as well as magnitudes. The biases can be severe enough to alter estimated directions of change as well as magnitudes. Economists studying growth and choice have begun to incorporate measures from other disciplines, but more needs to be done.

Economic choices are the result of complex evolutionary historical processes involving dozens of variables with ever-changing contexts. This paper will estimate two pooled-cross section multi-equation econometric systems model, and show that when economic models of growth and choice omit political, social, historical, and psychological factors they are subject to significant levels of specification error. When socio-political variables are included without incorporating mutual interactions, serious levels of simultaneity bias arise.

Keywords: complexity, sociopolitical, bias, evolution

TABLE OF CONTENTS

INTRODUCTION

1) Derivation of Equilibria
2) Theoretical Controversies
3) The Futility of Conventional Economic Theory
4) Bias Towards Rational Utility Maximization

EXAMPLES OF PROBLEMS CAUSED BY NARROW FOCUSES
1) Models of Economic Growth
2) Microeconomic Economic Behavior
3) Cultural History
4) Financial Dynamics
5) Estimates of Job Creation
6) Other Disciplines

SOLUTIONS

ESTIMATION

CROSS SECTION MODEL OF LDC'S

THE TIME SERIES MODEL OF BRITAIN 1895 - 1965

CONSUMER CHOICE

CONCLUSION

REFERENCES

APPENDIX

INTRODUCTION

A complex interdependent world requires different modeling strategies than are conventionally pursued by economists. In attempting to create compact theoretical models involving a relatively small number of variables which can be subject to differential calculus and simple hypothesis testing, economists necessarily omit critical facts which can significantly alter conclusions.

Because interactions among and between economic, political, social, and psychological variables are so significant, far reaching, complex, and dependent upon specific historical contexts; conventional utility maximization theory, OLS regressions, or simple two-equation systems inherently offer severely misleading answers with respect to questions regarding economic choice, economic development, and optimal economic policy. Economic choices are the result of complex evolutionary historical processes involving dozens of variables with ever-changing contexts. To draw conclusions based upon theoretical utility maximization analyses is like trying to crack the "RSA 129" data security code with a pencil and paper rather than a network of supercomputers.

Variables often exercise effects upon other parameters derivatively through a long chain of intermediate variables. Omission of these linkages can bias results significantly--rendering major influences to obscurity, and elevating minor factors to unmerited heights.

In addition, the effects of some variables may remain latent for decades until they reach certain thresholds, or until other variables reach critical levels--and then have explosive effects. Upon reaching thresholds these variables can exercise significant effects for a few years, and then disappear again. Their effects can irrevocably alter the course of growth or economic choices for decades--as opportunities are missed or capitalized upon.

This paper will show that when economic models of growth and choice omit political, social, historical, and psychological factors they are subject to significant levels of specification error. When socio-political variables are included without incorporating mutual interactions, serious levels of simultaneity bias arise. The biases can be severe enough to alter estimated directions of change as well as magnitudes.

Economists attempt to create artificial beginnings and endings in a world where there are none, and vainly seek to impose order on ever changing processes. Explanations and forecasts by M.D.'s regarding the behavior of our arms and legs cannot be made on the basis of a look at these appendages alone, without examining the brain, blood flows, nutritional inputs and other factors.

In If You're So Smart: The Narrative of Economic Expertise, (which also ridicules some of the models we support below) Donald McCloskey (1990) summarizes the dilemma of the "modernist" economist, who asserts without looking at the history of science that only her method, via ac ratio, yields clear and distinct ideas. But the range of the ratio is too narrow, though clear and distinct within it. It resembles the method of the drunk who looks for his keys under the lamppost because the light is better there. The extreme explicitness of modernist reasoning under the lamppost is accompanied by extreme vagueness outside its range.

The need to create easily tractable models has led economists to waste time, energy, and paper in many fruitless pursuits, including 1) the derivation of equilibria in an ever-changing world, 2) pointless theoretical controversies, and 3) the development of theoretical answers applicable only to fictional worlds. It has also biased work and results in favor of research utilizing the assumption of rational utility maximization.

1) Derivation of Equilibria

To derive equilibria in the small tractable models favored by economists, may be fulfilling to mathematically inclined individuals--but with respect to representing reality, predicting economic outcomes, and prescribing policies, it is a waste of time--because in modern societies changes in one of the multitude of variables inevitably omitted from basic models will with probability .999 shift the economy from the derived equilibria. Policy makers and business people have never and will never be concerned with the existence of equilibria, because in the real world equilibria do not occur. Whether or not economies fall into equilibria, or have multiple equilibria "on paper" in limited hypothetical worlds simply does not matter.

2) Theoretical Controversies

Battles fought in economics between well-specified tightly packaged theories also tend to be fruitless. Rational expectations theorists and their opponents continually attempt to refute the positions of opponents with ever more elaborate mathematical models. However, in reality, there are elements of truth in both sides of this controversy, and in many other disputes. The key question is: to what extent, and in which situations, are these positions correct. Estimation of comprehensive models incorporating each of these theoretical truths on a wide range of data covering a variety of contexts is needed to determine the degree to which each of the alternative constructs is true. Theoretical elimination of opposing truths is generally pointless.

3) The Futility of Conventional Economic Theory

Economic theories seek to derive cause and effect relationships, and to determine the direction of change. They ultimately seek to find whether variable A causes variable B to go up or down, or to stay unchanged, if the results are indeterminate, or if changes in other variables may cause the direction of change to be altered. The changes depend on the assumptions, and the contexts presented in the theoretical model.

Theory can usefully pave the way for sensible empirical efforts, display the logic of arguments, delineate plausible paths for causes and effects, clarify the roles of variables, and provide answers applicable to limited time frames and geographical boundaries--but it cannot answer any ultimate questions.

New contexts continually change the strength and direction of results. The nature of the contexts depends necessarily upon broad empirical questions which can only be resolved through broad empirical searches. The elaborate mathematical apparatuses which are used to determine whether something goes up or down on paper are relatively meaningless because the answers depend upon contexts--which are empirical, too complex for extant theoretical tools to analyze, and ever changing.

It is impossible to determine the direction of change theoretically in a world where hundreds of factors interact with each other. Even if a computer could mathematically calculate a direction of change in such a world, the result would be meaningless--dependent upon hundreds of assumptions which should be subjected to empirical verification.

Bias Towards Rational Utility Maximization

Finally, the emphasis upon creating pseudo-scientific tightly specified theoretical models leads to a self-fulfilling dependence upon models assuming rational utility maximization. Economists will generally not even look at more complex behavioral models because such models cannot be tightly specified with a limited number of variables, and do not produce mathematically elegant equilibria. Alternative models can be based on dozens of variables and several stages. Therefore, only tight models derived from rational utility maximization are admitted into the game. However, according to bounded rationality, Lancastrian Demand, and psychological theories, choice is a less tractable process involving a multitude of variables and historical contexts.

CONSEQUENCES OF NARROW FOCUSES

Economic growth models have evolved through time from simple production functions, whereby output is a function of capital and labor inputs. Many other factors have been added, most notably education and technical progress. Recently researchers have added a number of socio-political variables. However, growth is dependent upon dozens or hundreds of critical factors, some of which operate through other inputs, which are, in turn, influenced by growth. Omitting linkages may produce serious biases. We will list below some reasons why simple models based upon production functions will necessarily fail to explain growth and produce misleading coefficients. Although the foregoing illustrations may appear trivial, the nearly universal failure to incorporate complex interactions can produce nontrivial errors.

a) Money and Growth

Although some recent models have remedied past deficiencies, many well specified growth models have failed to include the monetary sector. Yet, monetary/exchange rate policy played a significant, and possibly determinate role in depressing British growth during the 1920's, and American growth in the thirties. By accommodating inflation in the 1970's, monetary authorities in the US and the UK clearly altered growth rates in that decade, and changed political and economic outcomes in subsequent decades. The weaknesses of financial institutions, or tendencies towards inflationary finance, play a critical role in depressing the economies of many developing countries, including some that possess high levels of physical and human capital accumulation (see Cameron (1972), McKinnon (1973), Kelley and Williamson (1974), Sylla (1977), Jung (1986), and Kindleberger (1987). The effects of financial weaknesses upon growth, might in some cases, not be visible directly, but only through low levels of capital investment or other channels.

Financial weaknesses, in turn, have multiple causes. Pressure originating in one of these causes, say fiscal irresponsibility, can reach a threshold, causing a breakdown in the financial system, a decline in investment, and reduced growth, etc. Without some focus upon the determinants of fiscal irresponsibility and upon financial strength, turning points will be missed in time series studies, and outliers will arise in cross section studies. Yet, regressing growth upon variables including measures of financial strength which omit the channels of influence--through factor shares--may produce insignificant coefficients. To understand British, American, or Latin American growth, economists must study the interconnected processes set forth above rather than build simple universal models or one to two equation regressions.

b) Unionization and Growth

Similarly, a regression of unionization rates on GNP using quarterly data between 1965 and 1985 in Britain would probably not produce significant results because too many other variables are influencing events. However, unionization may have played a significant role in determining wage rates, productivity, investor confidence, and the development of the Labor Party, which in turn affected government policies, social spending, tax rates, and investor confidence--which then affected growth. The effects can remain latent for years, and then explode in an event like the General Strike of 1926, or in the elections of 1974 and 1979, which probably altered investor confidence for years thereafter.

c) Ethnic Hatred and Growth

The influences of ethnic heterogeneity and attitudes to group differences have produced devastating consequences in Yugoslavia and elsewhere after a variety of political variables reached certain thresholds. Few, if any, sectors of this planet have been left untouched by ethnic violence. The Middle East, the Soviet Union, Northern Ireland, the Basque territories in Spain, Indonesia, Ethiopia, Bulgaria, Malaysia, Tibet, India, New York City, Los Angeles, and other areas all suffer from serious ethnic conflicts. These conflicts depress growth and divert resources towards weapons expenditures.

Predictions derived from simple models of growth as a function of labor, capital, and education are vacuous in the face of the forces unleashed when ethnic hatred surfaces.

Cognitive factors such as entrepreneurial vigor, the ability to accept and assimilate change, and the tolerance of other ethnic groups will play a greater role in determining the fate of democracy and capitalism in formerly Communist countries than such economic variables as income and price elasticities, rational expectations, or subgame perfect equilibria.

d) Political Instability and Growth

In recent years many economists have studied the effects of political instability upon growth. Many found that political instability did not significantly affect growth. However, political stability has been found by some (including Benhabib and Spiegel, 1992) to influence investment, which determines capital stocks, which, in turn, determines growth. Significant effects were also found by Alesina and Perotti (1995), Alesina, Ozler, Roubini and Swagel (1996), and Robert Barro and Sala I. Martin (1995).

e) Educational Investments and Growth

Educational investments in Eastern Europe or elsewhere, may not have boosted growth in the absence of the political and economic freedoms necessary to utilize knowledge. Educational investments without adequate physical infrastructure might also prove futile. Studies might thus underestimate the benefits of education when freedom and adequate infrastructure are present.

f) Income Distribution and Growth

The role of additional variables and contexts can be so strong that the signs of estimated effects can be reversed in different contexts. Thus, unequal income distribution may raise growth in certain contexts, lead to destabilizing violence in others, and remain latent in yet other situations. Ram (1988), Camano and Salvatore (1988), Alesina and Perotti (1995) have found irregular relationships between growth and income, and that other variables can alter the relationship.

Dozens of additional and no less critical influences upon growth could be cited.

ESTIMATION

In order to remedy the above mentioned deficiencies we have estimated two models. The first is a five equation model of Less Developed Countries, and the second is a 21 equation model of the United Kingdom from 1895 to 1980.

CROSS SECTION MODEL OF LDC'S

The simple cross section model below contains five endogenous variables, two of which (GNP Growth and Investment) can be classified as "economic", and three (Educational Investments, Political Stability, and Liberty) as "sociopolitical". Results are preliminary. There are five exogenous variables--labor supply, money supply, an index of religious conservatism, and westernization. Data is obtained from the World Bank (1989), Banks (1971), Adelman and Morris (1968 and 1973), and Muller and Volker (1988).

FLOW CHART

The model produced the following results:

DLRGDP = + 2.34 + .53DLK + .33DLL + .12ALH - .25LGDP65PW

(7.25) (11.01) (2.07) (1.85) (6.76)

- .03POLSTB -.03CIVLIB

(2.31) (1.40)

where DLRGDP = Change in the log of real GDP
DLK = Change in the log of Capital Stock
DLL = Change in the log of the Labor Force
ALH = Average level of education of Labor Force
LGDP65PW = Log of GDP per worker
POLSTB = Index of Political Instability
CIVLIB = Index of Civil Liberties (inverse)

and t statistics are in parentheses

IK85 = + .0826 - 1.22E-12K85 + .0020H85 + 2.056E-08L85

(9.27) (0.34) (1.45) (0.25)

- 0.0039POLSTB - .0007GVPGNP86 + .0003MONPGD86

(2.03) (2.68) (2.64)

where IK85 = Investment Rate (per $ of Capital Stock) 1985
K85 = Capital Stock 1985
H85 = Human Capital Stock 1985
DLL = Labor Force 1985
POLSTB = Index of Political Instability
GVPGNP86 = Government sector share of GNP 1986
MONPGD86 = Ratio of Money Supply to GDP 1986

ALH = - 0.44 - .19LGDPAVPW - .02LIB1 + .04RELCON66 + .10WESTRN66

(0.81) (3.34) (4.07) (1.65) (2.66)

where ALH = Average Level of Human Capital 1965, 1985
LGDPAVPW = Log of average level of GDP per worker 1965,1985
LIB1 = Index of Liberty
RELCON66 = Index of Religious Conservatism
WESTRN66 = Index of Westernization of Attitudes

POLSTB = - 1.78 - .90DLRGDP - .85ALH + 4.753E-09RGDP65B

(1.49) (1.97) (2.11) (1.87)

- 1.24E-18RGDP65B² - 1.18CIVLIB - .11CIVLIB²

(1.09) (2.35) (1.68)

where POLSTB = Index of Political Instability
DLRGDP = Change in the log of real GDP
ALH = Average Level of Human Capital 1965, 1985
RGDP65B = Real GDP 1965
RGDP65B² = Real GDP 1965 Squared
CIVLIB = Index of Civil Liberties (inverse)
CIVLIB² = Index of Civil Liberties (inverse) squared

CIVLIB = + 6.80 + .14POLSTB - 2.18ALH - 1.05E-09RGDP65B

(21.90) (1.98) (10.49) (2.11)

where CIVLIB = Index of Civil Liberties (inverse)
POLSTB = Index of Political Instability
ALH = Average Level of Human Capital 1965, 1985
RGDP65B = Real GDP 1965

H85 = + 1.09 - .91LH65 + .53LGDP65PW - .06LIB1 + .29WESTRN66

(0.61) (4.36) (2.70) (3.56) (2.23)

+ 0.006DURP6585

(0.32)

where H85 = Level of Human Capital 1985
LH65 = Log of Level of Human Capital 1965
LGDP65PW = Log of GDP (1965 prices) per worker
LIB1 = Index of Liberty
WESTRN66 = Index of Westernization of Attitudes
DURP6585 = Change in percentage of the population urbanized 1965 - 1985

The regressions show that westernization of attitudes and religious conservatism significantly affected education, and that education, in turn significantly affected output.

The model shows that OLS equations omitting the significant effects of political and social factors upon GNP will be biased. When instability is included, coefficients for the capital stock are 10% higher, and for the labor supply are 55% lower. No change was observed in the education coefficient. Root Mean Squared Errors (RMSE's) of forecasted income are 12.5% lower.

Similarly OLS equations incorporating political instability are biased because political stability, in turn, affects growth. In addition, political stability significantly influence investment, and Liberty, which, in turn affects growth, thus underestimating the significance of political stability. Therefore, simple OLS equations that include the effects of political instability in a simultaneous model render coefficients for capital stock 26.4% higher than simple OLS equations, for labor supply 30% lower, and for education nearly three times as high. Root Mean Squared Errors (RMSE's) of forecasted income are 6% lower.

The consequences of omitting socio-political factors, and of failing to incorporate interactions are summarized in the table below:

TABLE 1

CHANGES IN COEFFICIENTS (TENTATIVE RESULTS)

Policymakers expecting that an increase in educational investments of 10% will boost growth 20% may thus be gravely disappointed. Simple OLS tests of theories are meaningless given the degree to which "ancillary variables" and simultaneity can alter conclusions.

Omission of complex interactions and linkages thus produces serious biases.

THE TIME SERIES MODEL OF BRITAIN 1895 - 1965

Appendix B presents an Econometric Model of Political Social and Economic Change in Britain from 1895 to 1980.

The model heavily emphasizes economic variables. However, social, political, and cognitive variables play significant roles. Overviews of the model are presented in Table 1 which lists the sectors and endogenous variables, and in figures 1 and 2--flow charts of the model. Endogenous variables and their coefficients can be found in Appendix I. Twelve sectors are enumerated including (1) money, (2) prices, (3) production, (4) expenditure, (5) income, (6) labor supply and demand, (7) labor behavior, (8) social behavior, (9) population, (10) voting, (11) government policy, and (12) education. Key socio-political and cognitive variables include voting choices, government spending, social spending, education, strike activity, and unionization. Data is derived from Mitchell (1988), Feinstein (1972 and 1981), Craig (1970, 1972, 1973, 1974, and 1979), Williamson (1985), Sheppard (1971), Soltow, Hartwell (1972), Lydall (1955), Central Statistical Office (various a, b and c), and Butler and Sloman (1975).

FLOW CHART

Not available

The GDP, investment, and labor equations produced the following results in the cross section study of LDC's. Results are preliminary

D(LGDPFCEX)= -0.10 +0.52D(LALHRW) +0.57D(LKSTNTO) + 0.04LEDULTS

(2.79)(3.48) (10.67) (3.02)

- .10WAR2

(5.56)

where D(LGDPFCEX) = Change in log of nominal GDP
D(LALHRW) = Change in the log of All hours worker
D(LKSTNTO) = Change in the log of the Nominal Capital Stock
LEDULTS = Index of stock of education of Labor Force
WAR2 = Dummy variable for World Wars I and II

and t statistics are in parentheses

LINVKSTK = - 0.61 + 0.78LINVKSTK(-1) + 0.49D(LGDPF900)

(4.28) (15.24) (2.13)

- 0.04D(LCONSL,2) + 0.02GOLAMLIM - 0.01STDYSSTK

(0.42) (2.03) (1.90)

- 0.05IRE - 0.16WAR2

(1.95) (3.15)

where LINVSTK = Investment Rate (per $ of Capital Stock) D(LGDPF900) = Change in Real GDP
D(LCONSL,2) = Change in Long Term Interest (Consol) Rate Over Last 2 years
GOLAMLIM = Dummy Variable = 1 if government conservative -1 if liberal or labor
STDYSSTK = Strikes days lost in past 5 years
IRE = Dummy Variable = 1 if Southern Ireland part of U.K., 0 otherwise
WAR2 = Dummy variable for World Wars I and II

LHRSIMP = + 3.85 - 0.14LWRIRL + 0.35LLFPR1 + -0.05LWSKPW

(205.95) (2.87) (8.14) (4.33)

- 0.06LUNIONPW

(4.90)

where LHRSIMP = Log of Hours Worked
LWRIRL = Log of Real Wages
LLFPR1 = Log of Labor Force Participation Rate
LSWKPW = Stock of Average Wages per worker in past 5 years LUNIONPW = Log of Unionization Rate

LUNEMP= -2.40 +0.64LWRIRL -1.63D(LGDPF900,2) - 1.87LPFRATTO(-1)

(4.00) (10.98) (8.14) (4.33)

- 291.93OUTPERHR(-1) - 0.58WAR2

(4.43) (2.77)

where LUNEMP = Log of Unemployment Rate
LWRIRL = Log of Real Wages
D(LGDPF900,2) =Log of Change in Real GDP lagged 2 periods
LPFRATTO = Log of Profit Rate-profits/capital stock
OUTPERHR = Output per hour
WAR2 = Dummy variable for World Wars I and II

LLFPR1 = - 0.30 + 0.06LWRIND + 0.12LACMPTOT + 0.17LPFRATTO(-1)

(6.92) (6.67) (10.17) (5.23)

- 0.04LUNEMP - 0.01LWRIRLST + .02WAR

(8.50) (1.85) (1.71)

where LLFPR1 = Log of Labor Force Participation Rate
LWRIND = Log of Nominal Wages
LACMPTOT = Log of Actual/Potential GDP
LPFRATTO = Log of Profit Rate-profits/capital stock
LUNEMP = Log of Unemployment Rate
LWRIRLST = Log of Stock of Real Wages
WAR = Dummy variable for World Wars I and II

The model shows that OLS equations omitting the significant but indirect effects of political stability upon GNP through a simultaneous model will be biased as coefficients for the capital stock are 14% lower, for education are 25% lower, and for the labor supply are higher by 21%. Root Mean Squared Errors (RMSE's) of forecasted income are 18% higher when simultaneous interactions are omitted.

The consequences of omitting socio-political factors, and of failing to incorporate interactions are summarized in the table below:

TABLE 2

CHANGES IN COEFFICIENTS (TENTATIVE RESULTS)

CONCLUSION

Because interactions among and between economic, political, social, and psychological variables are so significant, far reaching, complex, and dependent upon specific historical contexts; conventional utility maximization theory, OLS regressions, or simple two-equation systems can offer severely misleading answers with respect to questions regarding economic choice and economic development. Economic choices are the result of complex evolutionary historical processes involving dozens of variables, not merely the maximization of utility with respect to income and prices.

A complex interdependent world requires different modeling strategies than are conventionally pursued by economists. In attempting to create tight, well specified models involving a relatively small number of variables that can be subject to differential calculus, economists omit key facts which can significantly alter conclusions.

This paper represents a quantitative return to modes of analysis pursued by classical economists in the 18th and 19th centuries, combined with the insights developed by systems theorists, post-war economists, and interdisciplinary researchers. We have presented two alternatives which appear to improve estimates.

REFERENCES

Abel, Andrew B. (1990), "Asset Prices under Habit Formation and Catching up with the Joneses", AEA Papers and Proceedings, May 1990, Volume 80 No. 2 38-42.

Acks, Kenneth (1993) "Cognitive Determinants of Human Capital Investments" unpublished.

Adelman, Irma, and Cynthia Taft Morris, (1988), Comparative Patterns of Economic Development 1850-1914. Baltimore: Johns Hopkins University Press.

Adelman, Irma, and Cynthia Taft Morris, (1973), Economic Growth and Social Equity in Developing Countries. Stanford: Stanford University Press.

Adelman, Irma, and Cynthia Taft Morris, (1968), Society, Politics, and Economic Development: A Quantitative Approach, 2d Ed. Baltimore: Johns Hopkins Press originally published in 1967.

Ahmad, Kabir V., (1983), "An Empirical Study of Politico-Economic Interaction in the United States," Review of Economics and Statistics. 65, 170-7.

Alesina, Alberto and Roberto Perotti (1995) "Income Distribution, Political Instability, and Investment" European Economic Review 92

Alesina, Alberto, Sure Ozler, Roubini and Swagel (1996) "Political Instability and Economic Growth" National Bureau of Economic Research Working Paper # 4173

Arthur, W. Brian (1989), "Competing Technologies, Increasing Returns and Lock-In by Historical Events," Economic Journal, XCIX (1989) 116-131.

Arthur, W. Brian (1991), "Designing Economic Agents that Act Like Human Agents: A Behavioral Approach to Bounded Rationality", AEA Papers and Proceedings, May 1991, Volume 81 No. 2. 353-359.

Banks, Arthur S., (1971), Cross Polity Time Series Data. Cambridge: Massachusetts Institute of Technology Press.

Barro, Robert and Sala I. Martin (1995) Economic Growth

British Labour Statistics, Historical Abstract, 1886-1968 (1971)

Brock, William A., and Stephen P. Magee, (1984), "The Invisible Foot and the Waste of Nations: Redistribution and Economic growth," in D.C. Colander (ed.) Neoclassical Political Economy. Cambridge: Ballinger Press pp. 177-85.

Butler, David, and Sloman, Anne, (1975), British Political Facts 1900-1975. London: Macmillan.

Camerer, Colin F., (1987), "Do Biases in Probability Judgment Matter in Markets? Experimental Evidence," American Economic Review, Volume 77, Number 5, (December), pp. 981-997.

Cameron, Rondo E., (1972), Banking and Economic Development: Some Lessons of History. New York: Oxford University Press.

Campano, Fred, and Dominick Salvatore, (1988), "Economic Development, Income Inequality and Kuznets' U-Shaped Hypothesis", Journal of Policy Modeling (10) 2 pp. 265-280.

Cappie, Forrest, and Alan Weber, A Monetary History of the United Kingdom, 1870-1982. London: George Allen & Unwin.

Central Statistical Office, Economic Trends. London: Her Majesty's Stationery Office

Central Statistical Office, National Income and Expenditure 1964-1974, London: Her Majesty's Stationery Office 1975.

Central Statistical Office, Financial Statistics. London: Her Majesty's Stationery Office

Central Statistical Office, Annual Abstract of Statistics. London: Her Majesty's Stationery Office

Cox, James C., and Seth Epstein, (1989), "Preference Reversal Without the Independence Axiom," Volume 79, Number 3, (June) pp. 408-426.

Craig, F.W.S., (1979), Minor Parties at British Parliamentary Elections: 1985-1973, London: MacMillan.

Craig, F.W.S., (1970), British Parliamentary Election Results: 1950-1970. London: MacMillan.

Craig, F.W.S., (1974), British Parliamentary Election Results: 1885-1918. London: MacMillan.

Craig, F.W.S., (1973), British Parliamentary Election Results: 1918-1949. London: MacMillan.

Craig, F.W.S., (1972), British Parliamentary Election Statistics: 1918-1970. London: MacMillan.

Craig, F.W.S., (1979), British Parliamentary Election Results: 1832-1885, London: MacMillan.

Faini, Riccardo, Patricia Annez, and Lance Taylor, (1984), "Defense Spending, Economic Structure, and Growth: Evidence Among Countries and Over Time", Economic Development and Cultural Change.

Fair, Ray C., (1978), "The Sensitivity of Fiscal Policy Effects to Assumptions about the Behavior of the Federal Reserve," Econometrica, Vol. 46, No. 5, (September), pp. 1165-1179.

Feinstein, Charles Hilliard, (1972), National Income, Expenditure, and Output of the United Kingdom, 1855-1965. Studies in the National Income and Expenditure of the United Kingdom, no, 6 Cambridge: Cambridge University Press.

Feinstein, Charles Hilliard, (1981), "Capital Accumulation and the Industrial Revolution." pp. 128-142 in Floud, Roderick, and McCloskey, Donald N. (eds.) 1981a The Economic History of Britain Since 1700. Vol. 1 1700-1860. Cambridge: Cambridge University Press.

Frey, Bruno S., (1978a), Modern Political Economy, New York: John Wiley & Sons.

Frey, Bruno S., (1984a), International Political Economics. Oxford: Basil Blackwell.

Frey, Bruno S., (1978b), "Political-Economic Models and Cycles", Journal of Political Economics, 9, pp. 203-221.

Frey, Bruno S., and Friedrich Schneider, (1981), "A Politico-Economic Model of the U.K.: New Estimates and Predictions," Economic Journal, 91, 737-40.

Frey, Bruno S., and Friedrich Schneider, (1978a), "An Empirical Study of Politico-Economic Interaction in the United States," Review of Economics and Statistics, 60, 174-83.

Frey, Bruno S., and Friedrich Schneider, (1978b), "A Politico-Economic Model of the United Kingdom," Economic Journal, 88, 243-253.

Gately, Dermot, (1980), "Individual Discount Rates and the Purchase and Utilization of Energy Using Durables: Comment," Bell Journal of Economics, 11, 373-374.

Goleman, Daniel (198 ), Vital Lies and Simple Truths. New York, Basic Books.

Grether, David M., and Charles R. Plott, (1979), "Economic Theory of Choice and the Preference Reversal Phenomenon." Volume 69, Number 4, (September), pp. 623-638.

Gupta, Dipak K., and Yiannis Venieris, (1981), "Introducing New Dimensions in Macro Models: The Sociopolitical and Institutional Environments," Economic Development and Cultural Change, 30, 31-58.

Hartwell, R. Maxwell (1972). "The Consequences of the Industrial Revolution In England for the Poor." In Hartwell et al., The Long Debate on Poverty: Eight Essays on Industrialization and "The Condition of England. IEA Readings, no. 9. London: Institute of Economic Affairs.

Hausman, Jerry, (1979), "Individual Discount Rates and the Purchase and Utilizations of Energy-Using Durables," Bell Journal of Economics, 10, 33-54.

Hibbs Jr., Douglas A. (1980) The Political Economy of Industrial Democracies, Cambridge, Mass, Harvard University Press.

Hibbs Jr., Douglas A., (1987), The American Political Economy: Macroeconomics and Electoral Politics. Cambridge: Harvard University Press. HB3743.H52 Bobst.

Hoch and Loewenstein (1990), "Time-inconsistent Preferences and Consumer Self-Control, "Journal of Consumer Research, Inc. Volume 17, March 1991 492-507

Jung, Woo S., (1986), "Financial Development and Economic Growth: International Evidence," Economic Development and Cultural Change.

Kahneman, Daniel (ed.), 1982, Judgment Under Uncertainty: Heuristics and Biases, Cambridge, New York, Cambridge university Press.

Kahneman, Daniel and Amos Tversky, (1979), "Prospect Theory: An Analysis of Decision Under Risk," Econometrica, (March), 47, 263-291.

Kahneman, Daniel and Amos Tversky, (1984), "Choices, Values, and Frames," American Psychologist, (April), 4, 341-350.

Kahneman, Daniel and Amos Tversky, (1981), "The Framing of Decision and the Psychology of Choice," Science, Volume 211, (January 30), pp. 453-458.

Kelley, Allen C., and Williamson, Jeffrey G., (1974), Lessons from Japanese Development: An Analytical Economic History, Chicago: University of Chicago Press.

Kindleberger, Charles P., (1987), Financial Deregulation and Economic Performance: An Attempt to Relate European Financial History to Current LDC Issues", Journal of Development Economics, 27, pp. 339-353.

Kindleberger, Charles P. (1989) Manias, Panics, and Crashes, New York: Basic Books

Kunreuther, Howard, and Paul Slovic, (1978), "Economics, Psychology and Protective Behavior: A Realistic Economics of the Consumer Requires Some Psychology,", Volume 68, Number 2, (May), pp. 64-69.

Kunreuther, Howard, et. al., (1978), Disaster Insurance Protection: Insurance or Federal Aid?, Washington, D.C:.

Kunreuther, Howard (1973), Recovery From Natural Disasters: Insurance or Federal Aid?, Washington, D.C:.

Landau, David, (1986), "Government and Economic Growth in the Less Developed Countries: An Empirical Study for 1960-80", Economic Development and Cultural Change.

Lichtenstein, S and Slovic, Paul, (1971), "Reversals of Preference Between Bids and Choices in Gambling Decisions," Journal of Experimental Psychology, 89, (July), pp. 46-55.

Lichtenstein, S and Slovic, Paul, (1973), "Response Induced Reversals of Preference in Gambling: An Extended Replication in Las Vegas," Journal of Experimental Psychology, 101, (November), pp. 16-20.

Loewenstein, George, and Richard H., Thaler, (1989), "Anomalies: Intertemporal Choice," Journal of Economic Perspectives, Volume 3, Number 4, Fall, Pages 181-193.

Loewenstein, George, (1987), "Anticipation and the Valuation of Delayed Consumption," Economic Journal, 97, 666-684.

Lydall, Harold, (1955), "Income, Savings and Net Worth," Oxford Bulletin of Statistics, 17. HC10.O9.

Marimon, Ramon, McGrattan. Ellen, and Sargent, Thomas J. (1989), Money as a Medium of Exchange in an Economy with Artificially Intelligent Agents, Santa Fe Institute A Working Paper of the SFI Economics Program.

McCloskey, Donald (1990) If You're So Smart: The Narrative of Economic Expertise Chicago: The University of Chicago Press

McKinnon, Ronald I., (1973), Money and Capital in Economic Development. Washington: Brookings Institution.

Mitchell, Brian R., (1988), Abstract of British Historical Statistics. Cambridge: Cambridge University Press.

Muller, Georg P., and Volker Borrschier, (1988), Comparative World Data.

Nelson, Richard and Winter, Sidney G., (1982), An Evolutionary Theory of Economic Change. Cambridge, Mass., The Belknap Press of Harvard University Press.

New York Times. "Study Finds Depression to Be Disabling to Many: Scientists Liken the Ailment to other Chronic Conditions." August 17, 198.

North, Douglass C. (1981). Structure and Change in Economic History. New York: W.W. Norton.

North, Douglass (1973). The Rise of the Western World: A New Economic History. Cambridge: Cambridge University Press.

O'Brien, Patrick Karl, and Engerman, Stanley Lewis (1981). "Changes in Income and Its Distribution during the Industrial Revolution." pp. 164-181 in Floud, Roderick, and McCloskey, Donald N. (eds.). The Economic History of Britain Since 1700. Vol. 1 1700-1860. Cambridge: Cambridge University Press. Vol. 2 1860 to the 1970's.

Ram, Rati, (1988), "Economic Development and Income Inequality: Further Evidence on the U-Curve Hypothesis", World Development, Vol. 16, No. 11, pp. 1371-1376.

Reilly, Robert John, (1982), "Preference Reversal: Further Evidence and Some Suggested Modifications in Experimental Design," American Economic Review, 72, (June) pp. 576-584.

Ruderman, Henry, Mark Levin and James McMahon, (1986), "Energy Efficiency Choice in the Purchase of Residential Appliances." In Kempton, Willett, and Max Neiman, (eds.), Energy Efficiency: Perspectives on Individual Behavior. Washington, D.C.: American Council for Energy Economy.

Scharfstein, D.S., and J.C. Stein (1990) "Herd Behavior and Investment," American Economic Review, LXXX 465-479.

Schneider, Friedrich, and Bruno S. Frey, (1988), "Politico-Economic Models of Macroeconomic Policy: A Review of the Empirical Evidence," in Thomas D. Willett (ed.) Political Business Cycles, Durham: Duke University Press, pp. 240-275.

Sheppard, David K. (1971). The Growth and Role of U.K. Financial Institutions, 1880-1967. London: Methuen.

Slovic, Paul, et. al., (1977), "Preference for Insuring Against Probable Small Losses: Implications for the Theory and Practice of Insurance," Journal of Risk Insurance, 44, (June), 237-258.

Slovic, Paul, and Sarah Lichtenstein, (1983), "Preference Reversals: A Broader Perspective," American Economic Review, 73, September, pp. 596-605.

Soltow, Lee, "Long Run Changes in British Income Inequality," in Floud, Roderick, Quantitative Economic History.

Sylla, Richard Eugene (1977). "Financial Intermediaries in Economic History: Quantitative Research on the Seminal Hypotheses of Lance Davis and Alexander Gershenkron." in Gallman, Robert E., (ed.), Research in Economic History suppl. 1 Recent Developments in the Study of Business and Economic History: Essays in Memory of Herman E. Kroos. pp. 55-80. Greenwich Conn.: JAI Press.

Thaler, Richard H., (1981), "Some Empirical Evidence on Dynamic Inconsistency," Economic Letters, 8, 201-207.

The Gallup Organization, Gallup Polls Public Opinion In Britain.

Tufte, Edward R., (1976), Political Control of the Economy. Princeton, N.J.: Princeton University Press.

United States Department of Commerce, Bureau of the Census, (1989), Statistical Abstract of the United States, Washington, D.C: U.S. Government Printing Office.

United States Department of Health and Human Services, National Institute of Mental Health, (1986), Schizophrenia: Questions and Answers. Rockville, Maryland.

United States Department of Health and Human Services, National Institute of Mental Health, (1986), Depressive Disorders: Treatments Bring New Hope. Rockville, Maryland.

Venieris, Yiannis P., and Dipak K. Gupta, (1983), "Sociopolitical and Economic Dimensions of Development: A Cross-Section Model", Economic Development and Cultural Change.

Wallis, John Joseph and Wallace E. Oates, (1988), "Does Economic Sclerosis Set in with Age? An Empirical Study of the Olson Hypothesis", Kyklos, 41 (3) pp. 397-417. H1.A15.

Williamson, Jeffrey G., (1985), Did British Capitalism Breed Inequality. Boston, Allen and Unwin.

Winston, Gordon C., (1980), "Addiction and Backsliding," Journal of Economic Behavior and Organization, Volume 15, (December), pp. 295-324.

Wold, Herman (1975). "Modelling in Complex Situations with Soft Information." Paper presented at the Third World Congress of the Econometrics Society, Toronto, August 21-26, 1975. Group Report no. 5. University of Goteberg, Department of Statistics.

World Bank, (1989), World Tables.

APPENDIX I

SECTOR ENDOGENOUS VARIABLES

1) Monetary Sector

1) Bank Rate

2) Long term consol (interest) rate

3) Money supply

4) Monetary velocity

(Money demand)

(Outstanding credit)

(Securities held at Bank of England)

(Private short term interest rates)

(Bank reserves)

2) Prices

1) Retail prices

(Agricultural Prices)

3) Production

1) Output (GNP)

2) Productivity

4) Expenditure

1) Consumption

2) Investment

3) Consumption patterns by industry

5) Income

1) Profits

2) Wage rate

3) Income distribution

6) Labor Supply & Demand 1) Unemployment

2) Labor force participation rate

3) Hours worked

(Female labor force participation rate)

7) Labor Behaviour

1) Strikes

2) Percentage of workforce unionized

8) Population

1) Birth rate

2) Death rate

(Marriage rate)

(Urbanization)

9) Social Behaviour

(achievement motivation)

(altruism)

10) Voting

1) Percent of votes for Labor Party

2) Percent of votes for Conservative Party

11) Government Policy

1) Government spending

2) Taxes

3) Trade Union Legislation

(actions on suffrage)

(local vs. national expenditure)

(types of expenditure i.e. defense, welfare

(Variables in parentheses have not yet been incorporated into the model)

1. MONETARY SECTOR

(1) LBNKRT = .668 - .400LUNEMP - 2.058LMNSPD + .964LBNKRT.L + 1.607LCP38D

(2.63) (2.05) (4.38) (25.95)

R² = .92 DW = 2.07

LBNKRT = Log of Bank Rate
LUNEMP = Log of Unemployment Rate
LMNSPD = Log of Change in Money Supply = Log(Money Supply/Money Supply_-1)
LBNKRT.L = Log of Bank Rate lagged over one period
LCP38D = Log of Change in Consumer Price Index = Log (CPI/CPI_-1)

(2) LCONSL = - .128 + .065LBNKRT.L + .457LCP38D + .853LCONSL.L + .049LY38 - .021WAR2

(1.12) (1.74) (2.68) (13.56) (1.42) (1.70)

R² = .94 DW = 1.67

LCONSL = Log of Consol Rate
LBNKRT.L = Log of Bank Rate lagged over one period
LCP38D = Log of Change in Consumer Price Index = Log (CPI/CPI_-1)
LY38 = Log of Real GNP (1938 Prices)
WAR2 = World War Dummy Variable

(3) LMNSUP = - 4.857 - .020LBNKRT + .572LBLD38 - .476LCP38D + 1.809LOGYBO - .189LUNEMP

(19.02) (0.44) (20.65) (1.76) (20.82) (9.35)

R² = .98 DW = 1.70

LMNSUP = Log of Money Supply
LBNKRT = Log of Bank Rate
LBLD38 = Log of Total Government Debt (National and Local in Constant Prices)
LCP38D = Log of Change in Consumer Price Index = Log (CPI/CPI_-1)
LOGYBO = Log of GDP (nominal)
LUNEMP = Log of Unemployment Rate

(3) LIQPRF = - .958 - .388LCONSL + .111LY38.L - .314LCP38D.L + .227LBLD38 - .595WAR

(5.51) (6.82) (1.77) (2.52) (10.04) (0.54)

R² = .98 DW = 1.70

LIQPRF = Liquidity Preference = Log(Money Supply/(Consumer Prices*Nominal GNP) = Inverse of Velocity
LCONSL = Log of Consol Rate
LCP38D.L = Log of Change in Consumer Price Index lagged = Log (CPI/CPI_-1)
LBLD38 = Log of Total Government Debt (National and Local in Constant Prices)
WAR = War Dummy Variable

2. PRICE SECTOR

(5) LCPI = + 1.530 + .110LCPI.L - .076LCPI.L_-1 + .076LMNSUP.L + .155LLFPR1 + .579LULC + .085LIMPRS + .013WAR2

(12.22) (3.98) (1.66) (3.94) (2.11) (12.25) (2.85) (2.32)

R² = .98 DW = 1.70

LCPI.L = Log of Consumer Prices
LCPI.L = Distributed Lag of Consumer Prices
LMNSUP.L = Log of Money Supply lagged one year
LLFPR1 = Log of Persons Engaged divided by Potential Labor Force
(Population over 15) a measure of demand and tightness of labor markets
LULC = Log of Unit Labor Costs
LIMPRS = Log of Import Prices
WAR2 = War Dummy Variable

3. PRODUCTION

(6) LOGY = - 1.847 + .685LALHRW + .00002LKSLSL + .369LEDUC

(1.89) (5.33) (5.47) (5.03)

R² = .83 DW = 1.71

LOGY = Log of Nominal GNP
LALHRW = Log of hours worked in year by all workers
LKSLSL = Log of Capital Stock divided by number of workers
LEDUC = Log of Education Input = Weighted average of governmental education expenditure, past percentage of population under 15 in school, and teachers per pupil

(7) PTVTYL = 64.55 + .034I38 + 37.717EDUC + 15.930WAR2

R² = .96 DW = 1.93

PTVTYL = Productivity of Labour = Output per Person Hour
I38 = Investment in 1938 Prices
EDUC = Education Input = Weighted average of governmental education expenditure, past percentage of population under 15 in school, and teachers per pupil

4. EXPENDITURE

(8) LC38PC = - .155 + .184LY38PC + .082LCB + .604LKSLPC + .370LRMSMP3 - .029WAR2

(3.29) (2.02) (2.19) (6.59) (3.79) (3.62)

R² = .97 DW = 1.82

LC38PC = Log of Consumption per capita (1938 Prices)
LY38PC = Log of Real GNP per capita (1938 Prices)
LCB = Log of Cash Balances = Money Supply in 1938 Prices
LRSMP3 = Log of Share of Wages in National Income
WAR2 = World War Dummy

(9) LI38 = - 2.032 + .371LI38.L - .718LI38.L_-1 - .133LCONSL + .925LYBOT + .135GOV - .117WAR

(6.95) (9.68) (5.96) (2.02) (7.92) (2.05) (6.73)

R² = .97 DW = 1.98

LI38 = Log of Real Investment (1938 Proces)
LI38.L = Distributed lag of Log of Real Investment (1938 Proces)
LCONSL = Log of Consol Rate
LYBOT = Log of Nominal GNP
GOV = 0 if Liberal or Labor Government, 1 if Conservative
WAR = World War Dummy Variable = 1 1914-1918, 1940-1945, 0 otherwise

(10) LIMPTSR = + 250.500 + .125Y38 - .006Y38_-1 - 197.57RELMPS

(1.60) (12.18) (0.11) (3.81)

- 376.219RELMPS_-1 + .046ER.L - 1,005.724DTY.TUK.L - 121.567WAR2

(3.24) (2.31) (6.54) (4.40)

R² = .77 DW = 1.91

IMPTSR = Real Imports
Y38 = Distributed lag of Real GNP
RELMPS = Distributed lag of Relative Prices = (Import Prices/Domestic CPI)
ER.L = Exchange Rate lagged one year (L/$)
DTY.TUK.L = Measure of Tariff Rates = Customs Duties Collected/Total Imports)
WAR2 = War Dummy

5. INCOME

(11) LWRIND = + .208 + .932LWRINL + .99LCPRDL + .495LLFPR1 + .095LUNNPC - .020WAR2

(3.65) (44.29) (4.15) (5.00) (4.28) (2.21)

R² = .99 DW = 1.89

LWRIND = Log of Wage Rate Index--Nominal
LWRINL = Log of Wage Rate Index lagged one period
LCPRDL = Log of Change in Consumer Prices lagged one period = Log(CPI/CPI_-1)
LLFPR1 = Log of Labor Force Participation Rate = Persons Engaged divided by potential Labor Force (Population over 15) a measure of demand and tightness of labor market
LUNNPC = Log of percentage of Labor Force in Unions
WAR2 = War Dummy Variable

(12) PROF38 = - 140.38 - 196.233ULC - 15.110UNEMP + .224Y38

(2.44) (2.16) (3.18) (11.98)

R² = .89 DW = 1.82

PROF338 = Profits 1938 Prices
ULC = Unit Labor Costs
UNEMP = Unemployment Rate
Y38 = Real GNP 1938 Prices

(13) YDIS = + .423 + .115WGESHR + .000001BGVX38 + .000005TAX38

(36.63) (6.91) (1.33) (2.15)

- .154PFRAT.L + .00006CPI + .000005Y38.L - .0003UNEMP - .0008WAR

(3.38) (6.35) (0.60) (2.67) (0.71)

R² = .89 DW = 1.82

YDIS = Income Distribution--Gini Coefficient most years particularly before 1945 imputed
WGESHR = Share of Wages in National Income
BGVX38 = Real Government Expenditure 1938 Prices
TAX38 = Real Government Taxation 1938 Prices
PFRAT.L = Profit Rate lagged 1 year = Profits/Capital Stock
CPI = Consumer Price Index
Y38.L = Real GNP lagged one year

5. LABOR SUPPLY AND DEMAND

(14) LHRIP7 = + 3.291 + .022LWRIRL + .244LLFPR1 - .098LWSKPW - .041LUNPCL

(288.33) (4.27) (5.71) (4.15) (5.89)

R² = .93 DW = 1.73

LHRIP7 = Log of Hours Worked per Worker per year
LWRIRL = Log of Real Wage Rate Index lagged one year
LLFPR1 = Log of Labor Force Participation Rate = Persons Engaged divided by potential Labor Force (Population over 15) a measure of demand and tightness of labor market
LWSKPW = Log of "Wage Stock" per Worker = the Sum of Wages in Past 5 years, a measure of economic freedom
LUNPCL = Log of percentage of Labor Force in Unions lagged one year

(15) LUNEMP = - 1.388 + .314LWRIRL - 3.007LACMPT - .318LUNNPC + .480LUNEMP.L - .354WAR2

(2.79) (3.63) (2.84) (1.73) (6.90) (5.69)

R² = .88 DW = 1.79

LUNEMP = Log of Unemployment Rate
LWRIRL = Log of Real Wage Rate Index lagged one year
LACMPT = Log of Deviation of Income from its Long Term Trend. A measure of demand or capacity utilization Ratio of actual to potential output
LPFRAT = Log of Profit Rate = log (Profits/Capital Stock)
LUNNPC = Log of percentage of Labor Force in Unions
LUNEMP.L = Log of Unemployment Rate lagged one year
WAR2 = War Dummy Variable

(16) LLFPR1 = - .024 - .044LWRIRL + .321LACMPT + .521LPFRAT - .032LUNEMP - .199LWSKPW + .007WAR2

(0.56) (7.17) (8.14) (1.95) (5.90) (6.62) (1.31)

R² = .88 DW = 1.87

LWRIRL = Log of Real Wage Rate Index lagged one year
LLFPR1 = Log of Labor Force Participation Rate = Persons Engaged divided by potential Labor Force (Population over 15) a measure of demand and tightness of labor market
LACMPT = Log of Deviation of Income from its Long Term Trend. A measure of demand or capacity utilization= Log of Real Ratio of actual to potential output
LPFRAT = Log of Profit Rate = log (Profits/Capital Stock) lagged one year
LUNEMP = Log of Unemployment Rate
LWSKPW = Log of "Wage Stock" per Worker = the Sum of Wages in Past 5 years, a measure of economic freedom
WAR2 = War Dummy Variable

6. LABOR BEHAVIOUR

(17) LUNNPC = - .686 + .888LUNPCL + .037DTDACT + .307LWGSTK + .295LPFRAT - .039LUNEMP + .040WBL1PS

(2.38) (25.02) (2.80) (2.90) (3.11) (2.65) (2.20)

R² = .88 DW = 1.87

LUNNPC = Log of percentage of Labor Force in Unions
LUNPCL = Log of percentage of Labor Force in Unions lagged 1 year
DTDACT = Dummy Variable for Trade Union Legislation = 1 for Trade Union and Disputes Act, -1 for 1927-1944, 0 otherwise
LWGSTK = Log of "Wage Stock" Sum of Wages in past 5 years. A measure of worker's economic freedom
LPFRAT = Log of Profit Rate = log (Profits/Capital Stock)
LUNEMP = Log of Unemployment Rate
WBL1PS = Inverse of the "well Being of Labor" in past = (Unemployment rate/Average Wage Rate) in past 6 years measures accumulated stock of grievances

(18) STKPW = - 1.583 + 1.798UNPCD2 + 1.294WBL1RC

(2.16) (2.70) (4.28)

+ .368WBL1PS - 3.161PFRAT + 7.912GENSTK + .287DTUACS + .167GOV + .205WAR2

(1.74) (0.50) (13.91) (2.46) (2.14) (0.92)

R² = .80 DW = 1.51

STKPW = Strike Days Lost Per Worker
UNPCD2 = Change in percentage of Labor Force in Unions lagged 2 periods, Represents desire of new unions to assert themselves
WBL1RC = Recent Change in "Inverse of Well Being of Labor" = (Unemployment Rate/Average Wage Rate) - (Unemployment Rate/Average Wage Rate lagged 1 year)
WBL1PS = Inverse of the "well Being of Labor" in past = (Unemployment rate/Average Wage Rate) in past 6 years measures accumulated stock of grievances
PFRAT = Profit Rate = Profits/Capital Stock
GENSTK = Dummy for General Strike of 1926. Strike days lost in 1926 were much greater than other years. Assumes that once level reaches certain point get "bandwagon" effect
DTUACS = Dummy Variable for Trade Union and Strike Legislation = 1 for Trade Union and Disputes Act (1906-1926) 1 for 1927-1944, 0 otherwise

7. POPULATION

(19) BR = - 124.878 - 107.968Y38PC + 321.261YDIS - 6.201EDUC - .299UNEMP + .277DR - 2.431WAR2

(5.36) (3.33) (6.39) (2.60) (3.61) (1.58) (2.86)

R² = .77 DW = 1.77

BR = Birth Rate
Y38PC = Real Income Per Capita
YDIS = Income Distribution--Gini Coefficient most years particularly before 1945 imputed
EDUC = Education Input = Weighted average of governmental education expenditure, past percentage of population under 15 in school, and teachers per pupil
UNEMP = Unemployment Rate
DR = Death Rate
WAR2 = World War Dummy Variable

(20) DR = + 3.748 - .0004BGVX38 + .757DR.L + .915WAR

(3.75) (3.17) (11.94) (3.02)

R² = .83 DW = 2.47

DR = Death Rate
BGVX38 = Total Real Government Expenditure, represents spending on sanitation, health care, the poor, . . .
DR.L = Death Rate lagged 1 year
WAR = World War Dummy Variable = 1 1915-1918, 1940-1945, 0 otherwise

8. VOTING

(21) CBYIMPIMPGAL = + 2.782 - .879DUNEMP...GOV + .146DUNEMP...GOV.L

(0.29) (3.39) (0.78)

- 1.053STKPW...GOV - .267STKPW...GOV.L + 65.819RSIMP3 - 3.485RSIMP3.L

(3.46) (0.78) (4.72) (0.12)

- 11.413DTXRPC...GOV - 1.159DTXRPC...GOV.L + .003DBGX38...GOV + .019DBGX38...GOV.L

(2.22) (0.31) (2.51) (0.13)

- .076YIP - 21.525UNONPC + .049PCCPI.L + 2.372EL - 1.639WAR2

(1.05) (5.43) (0.56) (2.74) (0.89)

R² = .66 DW = 1.99

CBYIMPIMPGAL = Percentage of Votes for Conservative Party Actual Totals in Election Years Gallup Polls for Post-1945 Election Years Average of 2 Imputations 1895-1945 Nonelection Years
1) Changes at Constant Rates
2) Changes in Byelection Voting from last General Election in same District
DUNEMP...GOV = Distributed Lag of Changes in Unemployment Rate Multiplied by +1 if Conservative Government -1 if Labour Government
STKPW...GOV = Distributed Lag of Changes in Strikes Per Worker Multiplied by +1 if Conservative Government -1 if Labour Government
RSIMP3 = Distributed Lag of Workers Share in National Income
DTXRPC...GOV = Distributed Lag of Changes in Tax Receipts Multiplied by +1 if Conservative Government -1 if Labour Government
DBGX38...GOV = Distributed Lag of Changes in Government Spending Multiplied by +1 if Conservative Government -1 if Labour Government
YIP = Years in Power
UNONPC = Percentage of Workforce Unionized
PCCPI.L = Change in Consumer Prices Lagged 1 Year
EL = Dummy Variable for Election Year
WAR2 = Dummy Variable for War Years

(22) LABBYIMP3GAL = + 30.413 - 85.309RSIMP3 + .130UNEMP..GLIBZER + 7.803DTXRPZ..GOVLAB

(3.25) (3.84) (1.88) (1.65)

+ 27.075UNONPC + 14.192PWARA + 7.880WBL1PS + 10.654FEMSUF + 7.081WAR

(2.02) (7.75) (4.00) (3.77) (4.70)

R² = .96 DW = 1.91

LABBYIMP3GAL = Percentage of Votes for Labour Party Actual Totals in Election Years Gallup Polls for Post-1945 Election Years Average of 2 Imputations 1895-1945 Nonelection years
1) Changes at Constant Rates given 2/3 of total weight since early election sample too small
2) Changes in Byelection Voting from last General Election in same District
RSIMP3 = Share of Wages in National Income
UNEMP...GLIBZER = Unemployment Rate Multiply by +1 if Conservative or Liberal Gov't -1 if Labour Government in power
DTXRPC...GOVL = Distributed Lag of Changes in Tax Receipts Multiply by +1 if Conservative or Liberal Gov't -1 if Labour Government in power
UNONPC = Percentage of Labor Force in Unions
PWARA = Dummy Variable Represents Attitudinal Shifts following Wars = 0 1895-1918, 1 1919-1945, 2 1946- (Peacock Wiseman Hypothesis)
WBL1PS = Inverse of the "well Being of Labor" in past = (Unemployment rate/Average Wage Rate) in past 6 years measures accumulated stock of grievances
FEMSUF = Dummy Variable representing Female Suffrage = .8 1921-1926, 1 thereafter
WAR = World War Dummy Variable = 1 1915-1918, 1940-1945, 0 otherwise

9. GOVERNMENT POLICY

(23) LBGX38 = - 1.064 - .972LACMPT + .017LACMPT.L

(1.53) (2.21) (0.02)

+ 1.202LY38 + .005LABBYIMP3GAL - .007CBYIMPIMPGAL + .128WAR

(5.68) (1.98) (3.20) (3.89)

R² = .87 DW = 1.47

LBGX38 = Log of Real Government Expenditure 1938 Prices
LACMPT = Log of Deviation of Income from its Long Term Trend Lagged 1 year A measure of demand or capacity utilization= Log of Real Ratio of actual to potential output
LABBYIMP3GAL = Percentage of Votes for Labour Party Actual Totals in Election Years Gallup Polls for Post-1945 Election Years Average of 2 Imputations 1895-1945 Nonelection Years
1) Changes at Constant Rates given 2/3 of total weight since early election sample too small
2) Changes in Byelection Voting from last General Election in same District
CBYIMPIMPGAL = Percentage of Votes for Conservative Party Actual Totals in Election Years Gallup Polls for Post-1945 Election Years Average of 2 Imputations 1895-1945 Nonelection Years
1) Changes at Constant Rates
2) Changes in Byelection Voting from last General Election in same District
WAR = World War Dummy Variable = 1 1915-1918, 1940-1945, 0 otherwise

LTAX38 = - 2.312 + .464LBLD.T.L2 + .660LY38 + .519LCPI + .421LBGX38 - .051WAR2

(4.72) (4.72) (3.35) (4.12) (5.14) (1.43)

R² = .92 DW = 1.68

LTAX38 = Log of Tax Receipts 1938 Prices
LBLD.T.L2 = Log of Ratio of Outstanding Government Debt to Taxes lagged 2 years
LY38 = Log of Real GNP
LCPI = Log of Consumer Price Index
LBGX38 = Log of Real Government Expenditure 1938 Prices
WAR2 = World War Dummy Variable

(24) DTUACS = + .522 - .010CBYIMPIMPGAL + .993DTUACS.L - .175STKPW.L - .045GOV - .037WAR2

(2.03) (1.68) (25.73) (7.66) (1.48) (0.47)

R² = .92 DW = 2.00

DTUACS = Dummy Variable for Trade Union and Strike Legislation = 1 for Trade Union and Disputes Act (1906-1926) -1 for 1927-1944, 0 otherwise
CBYIMPIMPGAL = Percentage of Votes for Conservative Party Actual Totals in Election Years Gallup Polls for Post-1945 Election Years Average of 2 Imputations 1895-1945 Nonelection Years
1) Changes at Constant Rates
2) Changes in Byelection Voting from last General Election in same District

DTUACS.L = Dummy Variable for Trade Union and Strike Legislation = +1 for Trade Union and Disputes Act (1906-1926) -1 for 1927-1944, 0 otherwise. Lagged 1 year
STKPW.L = Strikes Per Worker lagged 1 year
GOV = Dummy Variable for Party in Power = 1 if Conservative Government, 0 otherwise
WAR2 = World War Dummy Variable