Long-Term Growth and Standard of Living in Pre-Industrial Europe

The quantification of Europe's macroeconomic performance between the sixteenth and the eighteenth centuries has attracted considerable interest over the last decade or so. Essentially this can be divided into three types of exercises.

Table 8.1 Long-term growth of per capita output (constant prices), seventeenth and eighteenth centuries

Notes: The one for France is a lower bound estimate.

Sources: Blomme, Buyst, and van der Wee (1994); Malanima (1994); Yun (1994); van Zanden (1993); Snooks (1994); Topolski and Wysczenki cited by van Zanden (Chapter 7, this volume); Marczewski (1961).

perspective on the Early Modern period and hints at quite a different ‘pre-Industrial Revolution growth model' for the continent.¹

A second type of research has focused instead on the evolution of real wages (PPP—purchasing power parity—deflated)² for both skilled and unskilled urban labour. This follows a tradition of evaluating the movement of GDP per capita or of productivity using these indicators when direct information is unavailable. The claim is that over the long run a fairly good match can be obtained between proxy and proxied variables (Williamson 1995). The data gathered recently for seventeen major cities reveal noticeable parallels with the findings above (Allen 1998; Ozmucur and Pamuk 2002). A gentle upward trend is displayed for the London—Amsterdam region between the late sixteenth and late eighteenth centuries,³ and confirms its strong performance as revealed in Table 8.1. Stagnation or slow decline characterizes the situation in Naples, Antwerp, Florence, Milan, Madrid, Paris, and Strasbourg to the west and south. This reveals some discrepancies with the picture presented earlier, for example, Paris and Strasbourg, but otherwise strong parallels with it too. To the east, in a region that has received less attention in this field, the picture of stagnation/decline is similarly visible, in Leipzig, Vienna, Krakow, Warsaw, and Istanbul. The exceptions are Augsburg and Gdansk. The conclusion points again to a probably slight long-run increase in GDP per capita in the aggregate, encompassing both dynamic and stagnant/regressive situations across Europe, as well as periods of growth alternating with stagnation or even with contraction.

A third approach corroborates to some extent the outline of the long-term evolution of per capita GDP shown above by providing a league table for this variable around 1810.

at quite a distance, a by now clearly peripheral group composed of Iberia, Italy, and Poland, with roughly half the income level of the Anglo-Dutch ‘economic core' (van Zanden 1998).⁴

This revision of the long-run economic evolution of pre-industrial Europe, which allows for some measure of growth, is far from implausible, on both theoretical and empirical grounds. The former rests on the severe critique to which the previously and widely accepted ‘stagnationist’ approach to this question has been subjected (Persson 1988; Grantham 1999). According to these authors, the economies of this period were not locked into a stationary equilibrium from which escape was possible only in the case of some strong exogenous shock. Technological progress was indeed available in most fields of production if the right conditions were met. Perhaps the most important of these was demographic growth, which led to market broadening and to incentives for a greater degree of division of labour and regional specialization. This laid the foundations of slow but steady processes of learning-by-doing as the increase in the repetition of productive operations increased the probability of finding better ways of operating. Technical knowledge was thus increased and, given the right institutions, became cumulative, irreversible, and transferable to future generations.

Recent data on productivity and structural change reinforce this point of view and come from three different areas of research. The most important, given the primordial role of agriculture at the time, is the way in which the productivity of labour in this sector mirrored the path of GDP per capita. From the early sixteenth to the middle of the eighteenth centuries it rose in England by 100% and in the Netherlands by 40%, whereas elsewhere it was more or less constant (Allen 2000).

The rising trend of urbanization and its particular intensity in the more dynamic northwest is a second reason for endorsing the growth scenario we are considering, particularly as regards the Netherlands, which achieved a remarkable degree of city development by the seventeenth century (de Vries 1984). This conclusion is founded on the widely recognized fact that productivity at this time was considerably higher and more apt to increase in manufacturing and the services, the two pillars of the urban economy, than in agriculture (van Zanden 1998). It is also borne out by the generally higher per capita incomes associated with life in the towns and the cities, a fact which in turn was responsible, at least in part, for the rising inequality of income distribution in the areas of high urbanization (van Zanden 1995).

A third factor that is likely to have contributed to an upward income trend is the expansion of the labour input relative to the population that was characteristic of this period. This could have originated in several ways. One was the decline of real wages that van Zanden (1999) has pinpointed and which could have driven workers to make a compensatory larger effort in terms of days and hours in employment.

An alternative and not contradictory interpretation links this increased collective exertion to the occurrence of an ‘Industrious Revolution' in Europe. According to de Vries (1994), what drove households to engage to a greater extent in wage employment, and to a lesser one in domestic duties, was the appearance of new consumption opportunities offered by the market for goods.⁵ This stimulated new levels of consumerism that could only be achieved by means of these greater workloads.

Although the study of the pre-industrial standard of living is still in its early days, at first sight, an evaluation of its movement should not present great difficulties, given the evidence on real wages and GDP per capita adduced above. In fact, agreement on this subject seems somewhat elusive. On the one hand, the evidence marshalled by van Zanden (1999) points in the direction of steadily declining real wages in most parts of Europe, with two negative effects on welfare. The first was a reduction in per capita consumption of better quality foodstuffs, such as animal products. The second was a loss of utility caused by the poorer segments of the community having to work more, as noted above, in order to make up for the shortfall in real income. A contrary, ‘optimistic' perspective is based on several arguments. To begin with, the alternative PPP real wage estimate by Allen (1998) brings to light a more ambiguous portrayal of the situation. In some places, wages rose, in others they fell, and in still others they stagnated. In the second place, one may want to question how different the marginal utility of leisure was from zero for the mass of the lowly paid, underemployed, and/or seasonally employed workers, as the ‘pessimist' case requires, and therefore whether working more brought a net loss of utility to them. Finally, there is the claim of a ‘consumption revolution', which appears to have swept through eighteenth-century Europe and was probably present in some places already in the seventeenth. The improvement and enlargement of living quarters, the growing acquisition of higher quality durables, and the consumption of large amounts of exotic foodstuffs in such diverse places as England, France, Spain, Tuscany, and the Netherlands, all signal a picture of broad material improvement, across all classes and in both city and country, that goes very much against the idea of stagnation or regression in the standard of living.⁶

The ‘material consumption' approach to this emerging standard of living controversy is not the only methodological possibility but it is probably the most frequently used.

One distortion this causes is that it leaves out that sizeable fraction of the population that was too poor to be worth recording in this way because they possessed so few durable articles of consumption. Another is that it fails completely to take into account current consumption, thus ignoring the bulk of acquisitions made in the course of a person's lifetime. As a result, questions pertaining to the life cycle of individuals receive no attention either. Any conclusion drawn therefore can only reflect their welfare status at a very special moment of their existence and yields little knowledge concerning the population as a whole.

From the point of view of this chapter, the main defect of the ‘material consumption' approach, however, is its exclusive focus on material articles. Although the latter were indisputably the principal source of utility of the pre­industrial population, it is argued below that during this period there was a significant shift in personal expenditure away from material and towards immaterial goods. The centrepiece of this exercise is human capital formation, an indicator that is not usually included in standard of living studies and can be used in two ways. As an item of personal expenditure, it can serve to gauge the economic capacity to acquire, in the same manner in which the presence of mirrors, textiles, or sugar and coffee can. Alternatively, human capital generates a stream of utility of a kind that is not commonly taken into consideration, probably because of the problems of measurement it raises. During the period in question, like some material assets it conferred status and rank on its possessor and it could enhance the productive capacity of its bearer. But it was also a means to the fruition of individual non-material satisfaction such as comes from a greater knowledge and a better understanding of one's self, an enlarged ability to communicate with others, a richer religious experience, or the possibility of participation in public or community life. The present study is directed at the second of these perspectives. It constitutes therefore a response to the call to widen the discussion on living standards in the past by adding to the conventional purchasing power of private income approach broader aspects such as are proposed by the Human DevelopmentIndex (HDI) methodology (Crafts 1997).

3.