The Political Foundations Of Modern Economic Growth: England, 1540-1800

Gregory Clark,University of California-Davis

Introduction

Did political changes in Britain in the late 17th century create a stable property rights regime that established the pre-conditions for the Industrial Revolution? This paper shows that contrary to popular suppositions changes of political regime had no impact on the private economy in England between 1540 and 1800. Rates of return on private capital, for example, were uninfluenced by either the turmoil of the Civil War of 1642-8 or the institutional change of the Glorious Revolution in 1688. England achieved stable property rights at least 200 years before the first stirrings of the Industrial Revolution. For most of this period there was little economic growth. Stable institutions may be a necessary condition for economic growth, but English history shows they are not sufficient.

The Debate On Institutions And Growth

As we look across the contemporary world we see a clear association. Countries with high incomes per capita tend to be stable democracies with well defined property rights. Poor countries tend to be autocracies with periodic outbursts of violence, and with pervasive corruption, nepotism, and ill defined property rights. This modern association leads naturally to the questions: Are stable democratic institutions a necessary condition for modern economic growth? And are they a sufficient condition?

Political instability hurts the economy in a variety of ways. As factions vie for control of the state property rights are uncertain. Insecure regimes may suppress economic activity if the groups benefiting include those hostile to the regime. Insecure regimes may also undertax and be unable to provide valued public goods if those taxed refuse to comply because of their hostility to the regime. Driven by the short term needs of survival the insecure regime may plunder capital owners or innovators, depriving the economy of investment or innovation.

The secure but autocratic regime is seemingly more favorable to economic activity. Suppose the autocrat is purely self interested. He or she will then seek to maximize their own income. If the autocrat has untrammeled taxing power they will still want to tax in such a way as to ensure stable property rights and vigorous economic activity, since they collect a share of all income. The revenue maximizing autocrat will generally not, however, lead to as great an economic output as a stable democracy, since the autocrat cares only about government revenue and gives no value to the consumption of the populous. The autocrat will thus set tax rates at a higher level than the democracy which tries to maximize total consumption. This higher tax rate lowers output.

A corollary of the above views of how political institutions can impede economic growth is that the return on capital should be high in adverse regimes. If the force impeding investment and innovation is the fear of confiscation, people will only invest where the rate of return is high enough to compensate them for these risks. At the end of the English Civil war in 1650, for example, the victorious parliament sold most of the deposed King's estates. The perpetuities owned by the crown sold for on average implied rate of return of 11.2%, at a time when private perpetuities yielded about 5.5%. The return premium available to investors in Royal property reflected the political uncertainty that attached to these properties.

The England experience between 1600 and 1770 seems to many to lend weight to the causal association between a stable democratic politics and economic growth. The Industrial Revolution was preceded by the Glorious Revolution of 1688. In the Glorious Revolution a corrupt, autocratic monarchy was replaced by a political system where the Parliament controlled the monarch. This political system was remarkably stable. After 1689 there were no coups, and few attempted coups, but instead an unbroken line of elected governments. Between the Glorious Revolution and the Industrial Revolution there was widespread change in the British economy: the transport system was improved, a large scale conversion to a purely private agriculture was accelerated, new institutions of finance and commerce were put in place, and the government became a secure borrower whose debt was regarded as the safest asset in the economy.

It has been tempting to argue that the Glorious Revolution created the pre-conditions for the Industrial Revolution. Among those unable to resist this interpretation have been both Mancur Olson and Douglas North and Barry Weingast. Both argue that the Glorious Revolution in England marks the important shift towards private property rights secure from government confiscation.

The Political Background

Between 1540 and 1770 there were a number of periods when the England experienced political turmoil, internal warfare, and important changes of political regime. In the late 16th century the impending death of Elizabeth I created great political uncertainty. Since at least 1578 it was clear she would die childless, and the Tudor dynasty would end. There were five serious contenders for the throne, none of whom Elizabeth had the slightest affection for. Elizabeth increased the uncertainty by having an act passed imposing severe penalties on anyone making claims as to the royal succession.

Rates of return on capital should have risen for two reasons in the waning years of Elizabeth. The uncertainty of the succession meant there could be a bloody power struggle after Elizabeth died. The lack of a successor meant that Elizabeth could not credibly commit to any long term contract with her subjects, if we interpret monarchs as behaving like predators in the North and Weingast fashion. If she wanted to expropriate in the declining years of her reign, she could do so at little cost to herself.

The Stuart kings who followed Elizabeth had an unhappy relationship with the English Parliament. Between 1603 and 1688 there was an interminable struggle between the Crown and the Parliament over the powers of each, fueled in the years after 1660 by the Catholic sympathies of the monarchy and the Protestantism of the people. The Parliament controlled taxation, and it used this power to try and rein in the monarch. As a result the monarchy was short of funds and resorted to various illegal and semi-legal exactions and confiscations to raise revenue. For example, it allowed obsolete regulations limiting economic activity to remain in force, then fined the transgressors.

The conflict between King and parliament resulted in open warfare in the years 1639-40, 1642-46,1648, and 1651. Armies destroyed crops and requisitioning food. Some towns were sacked in the war, others were subjected to sieges. Then from 1649 to 1660 the country was subject to a Puritan control that was uncertain and vacillating, and gradually dissolved internally.

The monarchy was restored in 1660, but soon the old strains appeared. Charles had Catholic sympathies in a Protestant country where religion was an important political issue. His brother James, who became king in 1685, was a Catholic in a largely Protestant country at a time when Catholicism was feared as representing a danger to the English state. The policies of James resulted in widespread fear and disaffection. When James had a son in 1688, William of Orange invaded in collusion with English allies. James found little support and fled, and in 1689 the Parliament installed William and Mary as monarchs. Under the new constitutional order Parliament had control over the actions of the monarchy. This new order was the foundation of the modern British state.

The success of the Revolution was not immediately obvious. William had come to England in part because the Dutch needed to preserve England as an ally in their struggle to stay free from French hegemony. Thus from 1688 to 1695 the new regime was engaged in a War against France on the continent, and internal wars against the partisans of James in Ireland and Scotland who were supported by the French. Only in 1697 when William and Louis XIV made a peace treaty was it clear that the new political settlement was secure.

The new regime ushered in a host of political and administrative changes. In 1692 a Land Tax was imposed which provided a large new source of funds for the Government, and formed a relatively predictable exaction on property owners. Also in 1694 the Bank of England was formed as the principal lender to the government ushering in a host of financial developments now called "The Financial Revolution."

Political Regimes And The Rate Of Return

Did any of the supposedly important political events of 1580-1689 affect rates of return in the private capital market? This paper uses information from 4706 transactions recorded by the Charity Commission to examine the private capital market from 1540 to 1837.

To test whether political turmoil affected rates of return I look at three series - real property (land, houses, and tithes), rent charges, and bonds and mortgages. For real assets I first estimated the coefficients of the regression equation, RET = a + b1T + b2T2 + c1DX + c2DH + c3DT + d1DBUB + d2DOLD + d3DCIV + d4DINT + d5DSTU + d6DNEWREV + d7DREV + e (1) where the variables are defined in table 1. The regression formally tests if there was any change in interest rates in periods of turmoil or of regime changes. If politics mattered then the estimated values of d2 , d3 , d4 , and d5 should be large positives reflecting the periods of turmoil while either d6 or d7 should be large negatives (depending upon when people believed a new regime was established). I estimated a similar regression for rent charges. I further estimate the same equation for both types of returns with only two regime dummies, for the Civil War and for the years after 1696 when the new regime was established.

Table 2 shows the estimated values of the coefficients for each of the periods of interest. None of the political and military convulsions have any quantitatively significant effect on private capital markets. The estimated movement of interest rates in the periods of turmoil and uncertainty are mostly by fractions of a percent and are also mostly in the wrong direction. The Glorious Revolution, counted either as an event of 1689 or of 1697, is associated with at best an estimated decline of returns on land of .14% from the late Stuart period, and a decline of 0.40% for rent charges.

TABLE 1: DEFINITION OF VARIABLES IN REGRESSION EQUATIONS ________________________________________________________________________

TABLE 1: DEFINITION OF VARIABLES IN REGRESSION EQUATIONS

________________________________________________________________________

VARIABLE DEFINITION

________________________________________________________________________

T Year

DX Dummy equal to 1 when the return is expected (as in

will)

DH Dummy equal to 1 when the asset is a house

DT Dummy equal to 1 when the asset is a tithe right

DBUB Dummy equal to 1 in 1720 (South Sea Bubble)

DOLD Dummy equal to 1 in 1578-1602 (years of

Elizabeth's reign with clear absence of

successor)

DCIV Dummy equal to 1 in 1639-1648 (Civil War)

DINT Dummy equal to 1 in 1649-1659 (interregnum)

DSTU Dummy equal to 1 in 1670-1688 (last years of

Stuarts)

DNEWREV Dummy equal to 1 in 1689-1696 (early Glorious

Revolution)

DREV Dummy equal to 1 in 1697-1770 (Glorious

Revolution established)

________________________________________________________________________

________________________________________________________________________ TABLE 2: RATES OF RETURN AND POLITICAL CHANGES, 1540-1770 ________________________________________________________________________ Period Land, Houses, Rent Charges Tithes _______________________________________________________________________ 1720 (Bubble) -1.18** -1.19** 0.95 -0.96 (0.13) (0.13) (.56) (0.56) 1578-1602 -0.47* - -0.33 - (last years of ERI) (0.19) (0.21) 1639-48 -0.26 -0.16 -0.04 -0.01 (Civil War) (0.15) (0.14) (0.19) (0.18) 1649-59 -0.13 - 0.00 - (Interregnum) (0.18) (0.16) 1670-1688 -0.23 - 0.05 - (last years Stuarts) (0.15) (0.15) 1689-1696 -0.27 - -0.35 - (early Glorious Rev.)(0.22) (0.21) 1697-1770 -0.09 0.12 -0.25 -0.18 (late Glorious Rev.) (0.20) (0.14) (0.21) (0.15) N 936 936 746 746 R2 0.27 0.26 0.29 0.28 ________________________________________________________________________

Note: The numbers in parentheses are standard errors. ** indicates that the estimate is significantly different from 0 at the 1% level, and * significantly different from 0 at the 5% level.

In contrast to England we can look at the experience in the area of Zele in the southern Netherlands over the same period. The countryside there was subject to several long periods of destructive military campaigns and of uncertainty between 1550 and 1750. The first of these was between 1581 and 1607. In the years 1581-92 the struggle for Dutch independence was taking place mainly in Flanders. After 1585 most of Flanders was in Spanish hands but the Dutch continued to raid the countryside from then till 1607. There was also warfare in Flanders in the period 1672-97 during the wars of the Dutch and the Habsburgs against Louis XIV. There was further fighting in the War of the Spanish Succession. When the rate of return from land holding at Zele over roughly the same years 1550-1750 was estimated from the regression

RET = a + b1T + b2T2 + cDWAR1 + dDWAR2 + gDWAR3

(2) where DWAR1 is 1 for the years 1581-1607, DWAR2 is 1 for 1672-1797, and DWAR3 is 1 for 1701-13, all the periods of warfare were associated with much higher rates of return. The estimated increase in the rate of return for the years of the Spanish reconquest of Flanders is 1.6%, on a base rate of 3.4%. The second long period of warfare from 1672 to 1697 produced an increase in rates of return of 0.9% from a base rate of 4%. The third period, the War of the Spanish Succession, produced the smallest effect, an increase in rates of return of about 0.5% again on a base of 4%. In all three episodes the effects of warfare seem to have been much greater than any effect we observe from Civil strife or changes in regime that occurred in England.

The regression test used may not give the regime change of 1688 in England a fair test. For the test asks if there was a change in returns in 1689, or in 1697 when the new regime was finally secure. It might have taken 20-30 years to convince people that there was really a new political regime. Perhaps there was simply an acceleration in the decline in rates of return that was occurring before 1689, instead of a sudden decline. To test for this I estimate the expression RET = a + bT + cREVT + dDBUB + e (3) for the years 1660 (the Restoration of the Stuart monarchy) to 1729, where T is the year, REVT is 0 from 1660 till 1696, and thereafter REVT equals the year T, and DBUB is a dummy variable equal to 1 in the year of the South Sea Bubble. This was done for returns on land, for rent charges, and for bonds and mortgages combined. The coefficient on REVT measures how the trend in rates of return changed after 1697 as a result of the Glorious Revolution. In all cases the coefficient on T is negative showing that returns on all assets were falling before 1697. But in neither the case of real assets, nor of rent charges, nor of bond and mortgages was the estimated coefficient g significantly different from 0 either statistically or quantitatively. Thus there was no acceleration in the decline in returns after 1697. The Glorious Revolution leaves no trace on the path of rates of return in the English economy between 1660 and 1730. Rates of return were falling in the years 1660 to 1696, and they continued to fall at the same rate thereafter.

Another way we can measure the effect of politics on capital markets is to look at asset values, and in particular on the value of land. Asset values should rise sharply in stable periods and decline in unstable. This effect appears even more strongly than the effect on returns on land in the case of land in Zele. In some years in the war of independence from 1581 to 1592 real land prices fell to less than 6% of their value at the outset of the war. In the later war with Louis XIV land prices fell by more than 50% at their minimum. The average fall in prices in the war periods was respectively 83%, 30% and 11%.

Did English land prices fall in periods of political instability, and did they rise after the Glorious Revolution? To test for the effect of political events on land prices I looked at the regression equation predicting the logarithm of real land prices per acre in terms of land characteristics, a time trend, and political changes. That is, LOG(RENT) = a + b1T + b2T2 + cCHAR + d1DCIV + d2DREV + e (4) where CHARi are a set of land characteristics such as plot size, DCIV is a dummy variable for the Civil War period (1639-48), and DREV is a dummy for the Glorious Revolution (the years after 1688 or 1696). Controlling for the long time trends in land values the estimated effect of the Civil War is for land values to increase by 0.1%. The effect of the Glorious Revolution, dating that in 1697, is 9.7% decline in land values. This estimated coefficient is significantly less than 0 at the 5% level. The decline in land values is stronger if I take 1689 as the break point.

I can again ask whether the Glorious Revolution represented just a turning point in the trend of land prices. That is, did land prices gradually rise after 1689 or 1697? The answer is that in the period 1660 to 1730 then real land values were rising before the Glorious Revolution, and the rate of increase of land values was the same after the Glorious Revolution. There was a growth of land prices of 0.6%- 0.8% per year that predated the Glorious Revolution (taking either 1689 or 1697 as the "real" date of the Revolution) and that continued after the Revolution at almost exactly the same rate. On either way of treating the problem the Glorious Revolution had no effect on land prices.

Implications

The above results suggest that we cannot in any way use England in the period prior to the Industrial Revolution as an example of the importance of political stability to economic development. Stable property rights may have been a necessary condition for the Industrial Revolution, but since they existed in England and Wales for more than 200 years prior to the Industrial Revolution they were certainly not a sufficient condition. In looking for an explanation of the Industrial Revolution we must look for factors other than the emergence of stable private property rights.