Download as a PDF

4 downloads 94 Views 195KB Size Report
Oct 26, 2005 - such as Samuel Pepys in the 1660s. The problem is to derive the appropriate weights for consumption in calculating the efficiency with which ...
When Did Modern Growth Begin? Gregory Clark Wissenschaftskolleg zu Berlin and University of California Davis ([email protected]) Oct. 26, 2005 When did modern growth begin? We have typically answered this question by looking at the levels of output per capita or, recognizing the importance of population in determining this in early economies, at the level of efficiency (TFP) of the economy. These measures suggest, though not unambiguously, a structural break in England around 1800 from almost not TFP growth towards persistent TFP growth. This paper points out that TFP growth is not the only or the obvious way to measure innovation rates in an economy. And that on alternative measures 1800 does not stand out as a particularly unique date.

Introduction: Measuring the Rate of Technological Progress Modern growth springs overwhelmingly and exclusively from the expansion of measured Total Factor Productivity (TFP). Year by year the measured efficiency of modern economies, the output per unit of inputs, grows in ways that investments cannot directly explain. At a proximate level capital accumulation appears as another important growth source, but the expansion of the capital stock can mainly be directly traced to the effects of TFP advance on raising the marginal product of capital and inducing more investment. Where TFP growth ultimately comes from has been the subject of intense debate: knowledge investments, scale economies, human capital, animal spirits have all been promoted as the source. Its ultimate source remains mysterious. As figure 1 shows, at least for England, 1800 seems to be the break point in world history between the era where TFP growth, the fuel for the engine of modern growth, was

limited and sporadic, and the era where TFP growth was in abundant and continual supply. The measured efficiency of the economy is little greater in 1800 than it was on average 600 years before. But after then efficiency consistently increases decade by decade. This is often taken to imply that the rate of innovation, the rate of introduction of new ideas, increased in the world economy after 1800. The new growth theory indeed often explicitly models TFP as a function of idea production. If we were to consult the history of technology, however, or the history of science, of music, of literature we would see no such sharp discontinuity. Indeed that has created for people such as Joel Mokyr and others, who have promoted science as the key to modern growth, the problem of why the Scientific Revolution of the seventeenth century so long predated the Industrial Revolution. Similarly if we were to consult social historians they find periods such as the seventeenth and early eighteenth as one of profound changes in consumption patterns, particularly for the upper classes. Tables 1 and 2, for example, summarize major scientific, technical and artistic innovations in Europe between 1200 and 1760. This does not look like a static society as the TFP picture would suggest, but a society with some dynamism. Thus there is a mismatch between history as read from TFP measures and history as revealed by the details of technological and other innovations. Which is a better index of the rate of innovation in these societies? From its definition efficiency (TFP) growth at the national level is

g TFP = ∑ bi g TFPi i

where i indexes sub-sectors of the economy, and bi the value of output in each sector relative to the value of GDP. Because the output of some industries is the input of others

∑b

i

>1 .

i

Thus national TFP growth is just the weighted average of TFP growth in each sector, weighting by the importance of each item in the value of production. It is thus the markets evaluation of the rate of introduction of new technique, weighting by the £ votes of the consumers of various different goods. Goods consumed in the pre-industrial society by the mass of people get a lot of weight, as do goods that can be exported so that they can be a larger weight in GDP than in domestic consumption. The same idea introduced into one economy will have more of less impact on TFP growth depending on the income level, the land endowment, and trade patterns. This shows up in England in the Industrial Revolution period where we can measure TFP growth rates for England as a whole and for the north and the south. Figure 2 shows these annual TFP growth rates by decade. Because most of the production in the revolutionized industries lay in the north of the economy, and was exported elsewhere, the TFP growth rate there rises quickly to the modern level, while in the south there is little measured TFP growth. On this basis we would say that the north experienced the Industrial Revolution but the south did not. Why should we take this measure of TFP as an approximation to the rate of production of new techniques? If entrepreneurs were searching for new ideas in a profit maximizing way this would be, perhaps, an appropriate formula to measure their success. Profit maximizing innovators who gained property rights in any new techniques they discover would consider both the likelihood of success for a given unit of cost in

searching in a given area and the economic value of any property rights. Thus research efforts, all other things equal, would concentrate on techniques which were used in industries producing a larger share of GDP. But in practice there is little sign that innovation, at least before the midnineteenth century, was driven in this profit oriented way. Property rights in ideas were typically of very little value in societies such as England and cannot in themselves have had much hand in the production of new techniques. If new techniques were introduced for other reasons – intellectual curiosity, military conquest, national pride, royal sponsorship, religious devotion – then using TFP as a way of summarizing the rate of advance of technique in a society may make little sense. Thus the introduction of firearms in the late middle ages was a major technological advance, and one with many long run implications for Europe. Yet conventional TFP measures would fail to detect any effect of this innovation since the costs of a unit of lethal force appears in no price index. Indeed, insofar as firearms forced an increase in expenditure on fortifications, since cannon soon made the medieval style of fortification useless and required massive new defensive walls, firearms would be associated with a decline in the measured outputs of societies per unit of input. Secondly, even if we decide that for economic purposes we are interested in measuring the importance of innovations by their economic impact, we would still have to decide what set of weights to use to measure this impact. The conventional measure uses as weights the share of each sector in output. From a welfare perspective, if we have an economy like England that becomes very open to trade, then a better measure might be the weights of each sector in consumption. In the Industrial Revolution era this

significantly reduces the measured rate of TFP advance, since the industries with rapid TFP advance such as cotton textiles tended to export their output. But different groups within society had very different consumption bundles depending on their income levels, and their occupations. The poorest in English society, the rural laborers, had a consumption pattern which was very simple and involved a very small range of goods: staple foods were the majority of the budget, with important weights also for heat, housing, and linen and woolen clothing. The weights of imported spices, sugar, wines, tea and coffee, chocolate, tobacco, silk, wax candles, books, household decorations, firearms, cutlery, glassware, travel, painting and decoration, were all zero or trivially small until the nineteenth century. The very rich, particularly in urban locations such as London, had consumption patterns by the late seventeenth century where all these goods were important items of expenditure. Studies of probate inventories, which detailed the goods bequeathed by people at time of death in England in the 17th and early eighteenth centuries, show a wide range of consumption patterns depending on income, location and occupation. Table 3, for example, details the numbers of inventories in England c. 1700 leaving certain types of household goods. Even though the husbandmen and laborers represented here will be a select group of their class with higher than average stocks of possessions, the difference between their consumption patterns and that of the gentry and the merchant class is clear. Since rural laborers constituted the mass of consumers in England before 1800 weighting efficiency advance by national consumption gives a heavy weight to their consumption, even when we reckon on their much lower incomes. Thus we can construct, as is done below, measures of the rate of TFP advance from the perspective of

rural laborers, or university professors, from 1200 to 1869, and find very different patterns of TFP advance. Which weighting best describes the rate of technological advance in pre-industrial England? The way I measure rates of technological advance for various goods is by simplification of the following approximation for the efficiency of production of individual goods. In a competitive market for each product, i, the price will be determined approximately by, α

pi =

α

β

r i pk i w i s

γi

Ai

where r is the return on capital, pi the price of capital goods, w the wage, s land rents, A the TFP of the sector, and α, β, and γ the shares of each input in costs for good i. The numerator of the above fraction is essentially a measure of the “cost of production” of each good. If prices fall relative to the cost of production, the efficiency of production for this good has improved. In practice determining the weights α, β, and γ would be very labor intensive and perhaps not possible. So to simplify this I construct for the economy in the years 1200-1869 by decade a standardized cost of production α

C = r α pk wβ sγ where α, β, and γ are the average cost shares for England over time. In practice costs shares for most final products would not deviate two sharply from the economy wide average. Then I measure technological advance in any sector i in period t as

Ait =

Ct pit

Thus the rate of technological advance experienced by different classes of consumers will be A jt =

Ct Pjt

where j indexes different classes of consumers, and Pjt is the price index for their consumption, measured as

Pjt = ∏ pit

bij

i

where bij is the expenditure share of consumer j on product i. Using this measure how does efficiency advance in England 1200-1869 appear from the perspective of different groups within the economy.

Rural Laborers The lowest income group in the society for most of this period is rural laborers. Figure 3 shows the real wage of these workers from the 1200s to the 1860s. There is clearly little gain for this group in long term wages before 1800. Indeed by the 1860s their wages still did not equal those of the 1450s at their peak in the late middle ages. But real wages before 1800 are confounded as a measure of efficiency advance by important changes in population size that took place between 1200 and 1800. Figure 4 shows the preferred measure, which is the economy wide costs index divided by the average price of the laborers consumption packet. Now we see much less variation in the years before 1800 (though still some that is inversely correlated with population, which is surprising) but a very sharp takeoff in economic efficiency as measured from the rural laborer’s perspective after 1810.

Consumers with More Modern Tastes. As table 4 shows the consumption basket of rural laborers contained very few goods, and most of these staple foods. Except for sugar and housing, most goods consumed by laborers did not have prices that deviated much from the economy wide cost index in the years before 1800. But there were good whose prices were declining relative to average prices, and relative to the general production cost. These tended to be goods consumed by those at the upper end of the income spectrum: we will consider below why there would be this bias. Table 5 lists these in descending order of estimated relative efficiency in production in 1700-59 (before the Industrial Revolution) compared to 1200-49. Pride of place in this list goes to books. Efficiency is estimated to have risen almost 40 fold following the introduction of the printing press in the 15th century. The next commodity to experience dramatic price declines was cane sugar. Sugar cane cultivation, long established in India and the Middle East, spread to southern Europe in the later middle ages. In this period it was imported into England by the ships of the north Italian cities. With the Portuguese occupation of the Madeira’s (1419) and the Christian conquest of southern Spain and the Canary Islands (1495) sugar cane cultivation spread to southern Spain and the Atlantic Islands. By the 1490s, before sugar colonies were established in the new world, the price of sugar in England had fallen to one fifth of its level pre 1350 (relative to general production costs). Further dramatic price declines came in the late 17th century with the establishment of sugar colonies, such as Jamaica, in the Caribbean using slave labor.

A number of spices imported from South and South East Asia also saw dramatic declines in prices relative to costs over these years. Pepper, the most important spice consumed in England, was produced mainly in Kerala, India until the nineteenth century, and its price declined to one fifth its real medieval level by the early eighteenth century, despite periods of very heavy taxation in these years. Figure 5 shows the details of that decline by decade. The next commodity in table 5 in terms of the extent of efficiency advance was gunpowder, though here the first observation is from 1379. After gunpowder comes glass in terms of price declines. Glass for glazing has prices quoted by the square foot from 1302 to 1869. This series was supplemented by the prices of glass bottles of standard sizes, and the prices of wine glasses. Paint throughout this period was composed of an oil base (typically linseed) combined with a pigment (sold by the lb.). The price series for pigments is the average of the price of red lead, white lead, Spanish white, verdigris, vermilion, Spanish black, yellow ochre, red ochre, ivory black, umber, and black lead. Clearly there was a dramatic decline in silk prices from the medieval period on, more significant than for the other textiles. This decline happened despite the imposition of various duties on silk imports in the seventeenth and eighteenth centuries with the aim of encouraging English manufacturers. The decline in silk prices obviously continued through until the 1860s. For in 1849 the Clothworkers’ Company in London introduced into the clothing of its male almsmen a “black silk neckerchief” (price 3/10 which is 46 d.). By 1852 the almsmens’ hats were also made of silk. Thus silk, which had been an article only for the very wealthy in the middle ages (1 oz of silk thread in the 1280s cost 7

days wages of a building laborer) had become by 1849 part of the clothing of those of quite modest means. But most of the price decline was accomplished before 1760. Table 5 reveals that there were significant improvements in the efficiency of production of a variety of goods in pre-industrial England, most of these being goods consumed in significant amounts only by the rich, people whose consumption patterns were not that dissimilar from ours. The consumer I have in mind here is not the aristocrat, but the professional such as ourselves – the doctor, the vicar (earlier the monk), the grammar school master, the university professor, or the high Civil Servant such as Samuel Pepys in the 1660s. The problem is to derive the appropriate weights for consumption in calculating the efficiency with which their consumption pattern was delivered. Budgets for such households in pre-industrial England are rare. Also the budgets for the noble households that do survive tend to aggregate the consumption of the masters with that of the servants, who received their wage partly in board and lodging. We do, however, get good estimates of the typical consumption patterns of modern consumers. The weights for this index are designed to echo those of the modern consumer, as inferred from the DEFRA, National Food Survey, 2000, and the Office of National Statistics, Family Expenditure Survey, 2000. Thus food, alcohol and tobacco gets a weight of 74.5% in the COL index of the poor, but only of 18% in the budgets of modern workers.1 For many modern expenditures, however, we have no direct equivalence (automobiles are a very important category of modern expenditure, for example). Thus at least at this stage the weights are

1

I included only a share of the costs of prepared meals and food eaten out of the home, for which a substantial part of the payment for such meals is for the labor in food preparation and service, and the rental of the cooking and eating establishments. The rest of the cost of these foodstuffs was included under “services” and “lodging” costs.

a very rough approximation. Table 6 reports the budget weights for modern consumers and the weights used here. Figure 6 shows the resulting relative cost of living for “modern” consumers versus farm workers. The overall effect of changing the shares of the consumption basket towards modern shares (which stand in for high income consumers in the past) is that living costs rose much more slowly in the years 1280 to 1650 for high income consumers than for farm workers. Indeed for the years before 1500 average living costs for the rich were double those of the poor relative to c. 1650. Thereafter if there was any trend it was towards somewhat slower increases in costs for the poor as opposed to the rich. In part this was because of the importance of wages in the cost of living of the rich, and the substantial rise in wages relative to other costs in the years after 1800. Finally I calculate a hypothetical efficiency of production for the rich (doctors, attorneys, professors) versus farm workers for 1280-1869. This is shown in figure 7. For consumers with modern tastes there is significant efficiency advance before 1800, and little sign of discontinuity at this date.

Conclusion The appearance of a sharp discontinuity in the rate of technological advance for European society around 1800 is very dependent on the weighting that is given to the various efficiency advances that were occurring before and after 1800. The early economy did not lack technological advances, it was just that most of these were in goods that did not appear in the consumption bundle of the average consumer: imported spices, sugar, books, gunpowder, paints, silk textiles, glass, and paper. And there are many

goods or services which were improved where we do not even have a price: clocks, music, theater, art, eyeglasses, and newspapers for example. TFP measures, since they weight efficiency advances by the value of production in the economy as a whole, thus give very little weight to these advances. But from the perspective of consumers with tastes like ours we would not in any way find England 1200-1800 to be in technological stasis. This still leaves several questions. The first is why real cost declines before 1800 tended to come overwhelmingly for goods consumed in quantity by a very small fraction of society? The second is whether the discussion above leads to a radical skepticism about ever defining the rate of technological advance in an economy? For there are all sorts of other weightings of goods in consumption that would produce very different patterns of efficiency advance over time.

Bibliography

Allen, Robert C. 2001. “The Great Divergence in European Wages and Prices from the Middle Ages to the first World War.” Explorations of Economic History, 38(1): 411448. Beveridge, Lord. 1939. Prices and Wages in England, Vol 1: The Mercantilist Era. Clark, Gregory. 2002. “Shelter from the Storm: Housing and the Industrial Revolution, 1550-1912” Journal of Economic History, 62(2) (June). Clark, Gregory. 2004a. “The Price History of English Agriculture, 1209-1914,” Research in Economic History, 22: 41-120. Clark, Gregory, ‘The Condition of the Working-Class in England, 1209-2004’. Forthcoming, Journal of Political Economy (2005), Dec. Clark, Gregory. 2005b. “The Long March of History: Male Farm Workers’ Wages and Living Standards in England, 1209-1869.” Working Paper, University of California, Davis.

de Vries, Jan. 1994. AThe Industrial Revolution and the Industrious Revolution.@ Journal of Economic History, 54(2): 249-70. de Vries, Jan. 1993. “Between Purchasing Power and the World of Goods: Understanding the Household Economy in Early Modern Europe,” in John Brewer and Roy Porter (eds.), Consumption and the World of Goods. London, Routledge. Dyer, Christopher. 1989. Standards of Living in the Later Middle Ages. Cambridge: Cambridge University Press. Hoffman, Philip T., David Jacks, Patricia A. Levin, and Peter H. Lindert. 2002. “Real Inequality in Europe since 1500.” Journal of Economic History, 62(2): 322-355. Latham, R. C. 1983. The Diary of Samuel Pepys, Volume X, Companion. Berkeley: University of California Press.

Shammas, Carole.

1990.

The Pre-industrial Consumer in England and America.

Oxford: Oxford University Press. Thorold Rogers, J. E. 1866. A History of Agriculture and Prices in England. Volume 2. Oxford: Clarendon Press. Thorold Rogers, J. E. 1888a. A History of Agriculture and Prices in England. Volume 3. Oxford: Clarendon Press. Thorold Rogers, J. E. 1888b. A History of Agriculture and Prices in England. Volume 6. Oxford: Clarendon Press.

Robinson, Howard. Britain’s Post Office: A History of Development from the Beginnings to the Present Day. Oxford: Oxford University Press. Weatherill, Lorna. 1993. “The Meaning of Consumer Behavior in Seventeenth- and Early Eighteenth-Century England,” in John Brewer and Roy Porter (eds.), Consumption and the World of Goods. London: Routledge.

Figure 1: Efficiency (TFP) in the very long run, England 1200-2000

Figure 2: Efficiency Growth Rates in the North and the South of England, 17701869

Notes: The north is composed of the counties of Cumberland, Northumberland, Westmorland, Lancashire, Durham, Yorkshire and Cheshire. The north had 30% of the population of England by 1851. Productivity growth declines in the 1860s in part because of the American Civil War which reduced the output of the cotton textile industry.

Figure 3: Real Agricultural and Builder’s Day Wages, 1209-1869

Source: Clark (2005a), Clark (2005b).

Figure 4: Production Efficiency in England from the Perspective of an Agricultural Laborer

Figure 5: Real Pepper Price, England by decade

2.00 1.80 1.60

Real Pepper Price-England

1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00 1200

1300

1400

1500

1600

1700

1800

Figure 6: The Relative Cost of Living, modern consumption versus farm workers (1700-59 = 1.00)

Figure 7: Efficiency of Production for Farm Workers and, Hypothetically, Modern Consumers (1860-9 = 100)

Table 1: Technological Achievements in Europe, 1200-1700 Date

Event or Discovery

Person

Place

c. 1200 1275 c. 1285

Windmills Gunpowder Mechanical Clock

St Albertus Magnus -

N. Europe Germany N. Europe

c. 1325 c. 1330 c. 1350

Cannon Crown Glass Spectacles

-

N. Europe France Venice

c. 1450 c. 1450 c. 1475 1492 1498

Printing Press Quadrant (ocean navigation) Musket Discovery of New World Sea route to India

Johann Gutenberg Columbus Vasco da Gama

Germany Italy, Germany Spain Portugal

1512 1522 1532 1544 1589

European Postal Service Circumnavigation of world Introduction of potato Introduction of tomato Knitting Frame

Franz von Taxis Magellan William Lee

Various Spain Spain Italy England

c. 1600 1600 1608 1614 c. 1620 1635 c. 1650 1654

Arabic numerals popularized Discovery of Electricity Telescope Logarithms Alarm Clock English Postal Service Mechanized silk spinning Liquid in glass thermometer

William Gilbert Hans Lipperhey John Napier

England England Netherlands Scotland England Italy Italy

1656 1665

Pendulum Clock Microscope

Ferdinand II of Tuscany Christiaan Huygens Robert Hooke

Netherlands England

1718

Silk Throwing Machinery

Thomas Lombe

England

Table 2: Achievements in the Arts in 1200-1700 Date

Event or Discovery

Person

Place

11201500

Gothic Architecture

-

N. France, England

c. 1315 c. 1350 c. 1390

The Divine Comedy Decameron Canterbury Tales

Dante Aligheri Giovanni Boccaccio Geoffrey Chaucer

Italy Florence England

c. 1400 1413

Harpsichord Perspective in Painting

Filippo Brunelleschi

Flanders Italy

1509 1576 c. 1587

Wallpaper Public Theater Tamburlaine the Great

James Burbage Christopher Marlowe

England England England

1600

Opera – “Euridice”

Florence

1602 1620 1637

Hamlet Newspaper Public Opera House

Jacopo Peri and Giulio Caccini William Shakespeare -

England England Venice

1709 1719

Pianoforte The Novel - “Robinson Crusoe”

Bartolommeo Cristofori Daniel Defoe

Italy England

Table 3: Percentage of Inventories with Given Objects, England, 1675-1725 Social Group

Number

Books (%)

Looking Silverware Pewter glasses (%) (%) (%)

Pictures (%)

Gentry Merchants and professionals

122 152

39 45

62 62

61 51

93 95

33 35

Husbandmen and laborers

360

4

16

2

89

0

Source: Weatherill (1993), p. 222.

Table 4: The Percentage of Expenditure by Category for Farm Labourers before 1869 Category of Expenditure

Food and Drink: Bread and flour Wheat Barley Oats and oatmeal Peas Potato Farineous Meat Fish Bacon Eggs Meat Milk Cheese Butter Dairy Sugar and Honey Beer Tea Coffee Drink Salt Other Food Housing Fuel Light Soap Light and Soap Services Tobacco Other (Clothing, Bed linen)

1787-96 (Horrell)

1840-54 (Horrell)

Assumed here

77.0 40.1 0.0 1.0 3.6 2.0 46.7 9.2 0.0 1.3 0.0 10.5 4.0 3.5 3.9 11.4 3.6 0.0 2.4 0.0 2.4 1.4

68.6 33.5 3.0 1.4 2.2 6.0 46.1 3.4 0.0 2.8 0.0 6.2 3.2 2.6 3.3 9.1 3.1 0.0 2.6 0.0 2.6 1.6

73.0 0.0 40.0 3.0 2.5 2.5 4.0 44.0 10.5 0.0 1.0 0.5 9.0 4.3 2.3 5.1 10.0 3.0 4.7 3.3 0.0 8.0 0.5 0.0

6.0 4.0 4.8 0.1 0.0 8.2

10.1 4.5 3.3 0.7 1.0 11.7

6.0 5.0 3.5 0.5 4.0 0.5 0.0 10.0

Sources: Horrell, ‘Home Demand’, pp. 568-9, 577.

Table 5: Commodities with Efficiency Gains in Production, 1200-1349 to 1700-59 Commodity

Books Sugar Ginger Gunpowder* Glasswares Cloves Pepper Nails Silk Thread Pigments House Rent Painting Oil Woolen Cloth Silverwares Tallow Candles Lamp Oil Paper* Wax Candles Cutting Tools Beer

Unit

100 Pages Lb. Lb. Lb. Sqr. Foot Lb. Lb. Lb. Oz Lb. House/week Gallon Yard Oz. Lb. Gallon Quire Lb. Axe Gallon

Notes: *Earlier prices from 1350-99.

Relative Production Efficiency (%) 39.4 14.9 10.1 8.06 5.39 4.62 4.46 4.43 4.17 3.95 3.63 2.90 2.65 2.58 2.58 2.55 2.17 2.12 1.91 1.85

Table 6: Budget shares for the two social groups before 1869 (% of expenditure) Category of Expenditure

All Food and Drink:

Farm Laborer (% expenditure)

Average Household, UK, 2000

Assumed here

74.5

18.0

26.00

5.4 0.4 6.0

24.0

23.0 1.5 3.0 3.0 1.5 3.0 35.0

5.0

2.0

3.0

Tallow candles Lamp Oil Wax candles/Gas Light

3.5 0.0 0.0 3.5

1.0

0.0 0.5 1.0 1.5

Soap

0.5

1.0

1.5

House Furnishing Textiles Silk Clothing Leather Goods

3.5

2.38

4.0

0.0 5.5 1.0

5.12 1.13

1.5 6.0 1.5

Paper Books/Newspapers/TV

0.0 0.0

1.0 4.0

1.4 5.6

Services

0.5

9.0

13.0

House Rent Pewter and Tin Wares Silverwares Glasswares Copper and Brasswares Paint Housing, Decoration Fuel

Sources: Farm Laborers, Clark (2004c), DEFRA, National Food Survey, 2000, Office of National Statistics, Family Expenditure Survey, 2000.