Economic Activity and Changes in Forest Cover:
Thoughts on New England
Over the course of the approximately 370 years since European settlement, the forest landscapes of New England have undergone a series of large scale transformations, affecting both area of forest cover and forest ecology. These changes were abundantly noticed in contemporary accounts, and have recently been documented, e.g. by Foster and his co-workers at the Harvard Forest, Petersham, MA (Foster 1992, 1995, Foster et al. 1992). They have showed that from a pre-settlement condition of at least 90% forest cover, the New England landscape was at first gradually, then quite rapidly denuded of forest cover to a low point of perhaps 20% around 1860. From that point on, there has been a gradual return of forest cover to the present status of about 80% land area.
This U-shaped curve of forest cover through time mirrors that observed in other areas of the world that have undergone a history of human settlement, extensive agricultural exploitation, and ensuing economic changes. Parts of Finland and Scandinavia, parts of the southern U. S. and Central America all show a similar U-shaped trend. Foster (1995) suggests that "[t]he New England landscape provides an opportunity to examine the long-lasting impact of extensive land use on forest ecosystems. Lessons learned from such an analysis may be applied towards understanding other parts of the world where forest cover has been substantially altered or removed and eventually recovered." However, there are many areas of the world which have followed very different trajectories in respect of forest cover under long-term human influence. The Mediterranean basin, for example, has never regained the forests it lost during many centuries of human habitation. Clearly, it is some particular aspects or dimensions of human economic activity which drives this trend, not simply economic activity in general. "Information on the economic, technological and historical factors underlying human activity provides a context for understanding the changing pressures on forests . . . " (Foster 1995).
It is the objective of this paper to examine several components of economic activity to make some inferences as to their relative importance in determining the historical course of the deforestation/afforestation curve in New England. This paper might constitute a first step in a much larger and more elaborate project to establish the possibility of defining indices of economic activity which could be used to predict deforestation/afforestation trajectories in other regions of the world. In the case of Central and South America, for instance, some areas are presently experiencing high and increasing rates of deforestation, while in other areas these rates may be slowing. Many areas are expected to reach the end of their unprotected, legally exploitable forest stock in the coming decades. Based on an examination of the economic systems in which the respective forests are enmeshed, might we predict the fate of forests in these areas over the next hundred years, say, during which period economic growth and development may be expected to continue unless interrupted by cataclysm?
For the present analysis, the examined components will be: the degree of "market embeddedness" of the rural economy; the development of transportation networks; extent and intensity of intra-regional vs. inter-regional trade; and, urbanization and industrialization processes. The time period examined will be 1750-1900.
Price Differentials and the Market Economy
A "market economy" is one in which prices for a good or service tend toward uniformity over a given region in space and time (Marshall, in Rothenburg 1992). Price differentials between small isolated markets motivate the movement of resources between markets, encouraging development of transport and communications networks. These flows tend to narrow price differentials as they smooth out regional availability of goods through negative feedback loops, and so to create larger markets. High transaction costs - including transport costs and others - tend to inhibit the development of markets, and themselves spur technological changes which lower costs. Community controls on prices also tend to inhibit the development of markets. There is evidence that at least up to 1750, formal and informal community controls over prices and wages - perhaps in tandem with high costs and difficulty of transport - were strong enough to slow or prevent the growth of regional markets (Rothenberg 1992). Community control was exerted both formally, by official price-setting at county and state levels, and informally, by the network of community trade relationships within and between townships and counties. After 1750, price differentials for commodities and labor increased for about 50 years, as the influence of communities over prices waned. Price variance peaked around 1800, then declined again, under the influence of developing regional markets. Thus, the evidence from price convergence tends to support the idea that the early stages of development of a market economy may be correlated with forest cover loss.
Expansion of the market economy from local to regional scale requires the free movement of goods from place to place. How severe were regional obstacles to the movement of goods in New England in the 18th and early 19th centuries? Boston and other coastal towns in New England lacked the large rivers which, in New York, Philadelphia, and Baltimore, facilitated the movement of goods to and from the hinterland. After the great success of the Erie Canal project in the Midwest, several small tidewater canals were built with the intent of improving interior access to the coast, but these did not prove successful for technical reasons. The Cumberland and Oxford in Maine, completed in 1827, was the most successful, but was only 20 miles long. It was supplemented by lake and river access, and survived competition with the railroad until the 1870's.
But before the advent of the railroads during the 1830's, movement of goods in New England was achieved primarily overland, on the networks of town roads. These roads were notoriously bad, being built by the respective towns primarily for the use of farmers in getting goods to local markets, and for getting people into town or to church, and were never designed for long distance use. Nevertheless, the frequency of long distance trips traveled by farmers between 1750 and 1850 for the explicit purpose of buying and selling goods, was quite surprising, and belied the ready notion that markets were merely local affairs during this period. Rothenberg, examining farmers' account books, found mention of 1,827 marketing trips during these 100 years, of a mean distance of more than 20 miles, and including many trips over 100 miles in length (Rothenberg 1992). Previous research, based partly on contemporary accounts (e.g. Robert Fulton in 1814, quoted in Taylor 1951) had assumed prices for overland transport of between 30-70 cents per ton-mile between 1800-1820, and falling prices thereafter, when there was a general price deflation in the U. S. and the era of turnpike-building commenced (see below) (Taylor 1951). Rothenberg's work with extrapolated wage rates shows, in contrast, that real transport prices may instead have ranged between 14 and 18 cents per ton-mile for the period of 1750-1850, and that despite the manifest obstacles to transport, farmers regularly moved goods over large distances to find better price opportunities. Goods thus moved included not only milk, cheese, butter, livestock and vegetables, but bulky freight such as bricks, shingles, hay, ashes, barrels, potatoes (Rothenberg 1992). Such extensive movements of farm products from place of origin to place of sale, often occurring during the summer and fall seasons when labor could least be spared from the farm, indicate a very considerable involvement of farmers in regional markets as early as the immediate post-Revolution period. This was at least a full human generation before the beginning of the so-called "transportation revolution". One of the first changes of this revolution was the building of turnpikes to supplement the network of country roads.
The turnpikes were straight roads, designed for ease of construction more than for ease of use. They were financed and built by private companies which then charged tolls for use. The period 1800-1830 has been called the "turnpike era" in the U. S. as a whole, but by as early as 1806, the peak years of turnpike building were already past in New England (Taylor 1951). Turnpikes varied widely in quality, but were only rarely a real improvement over the old roads, both because upkeep proved difficult and because of initial design flaws (Taylor, Rothernberg, etc). Indeed, of the farmer's account books examined by Rothenberg, only one included an entry for a turnpike toll. However, New Englanders were notorious for evading tolls, and the difficulty of actually collecting tolls was an important reason for the demise of most of the turnpike companies by 1835. The failure of these roads was due not simply to competition with the railroads, which began to be built on a large scale around 1830. Many of the turnpikes were being abandoned much earlier than the advent of rail - by 1819 in Massachussetts (Taylor 1951). Competition with water transport was not a big factor either, since turnpikes were built to complement river and canal routes. In New England, turnpikes lead primarily inland from Boston and other coastal cities. Examples of such turnpikes still in use are Route 9 through Newton and Framingham to Worcester, and Washington Street in Dorchester and Jamaica Plain. In a very few instances, such as the Middlesex Canal leading from Lowell to Boston, there was a protracted struggle between the teamsters (the professional wagoneers who owned teams of oxen) and the canals. The teamsters were able to charge punishingly high rates for the transport of goods from their points of origin on farms and factories, to the canal, and from canal to warehouse at the other end of the trip. The canals could only respond to these high costs by continually lowering the rates on the canal itself. As toll roads, however, most of the turnpikes were simply unprofitable. "Very few indeed returned any part of the original capital to the investors. Even in New England, where they were most successful, only 5 or 6 out of turnpikes paid barely satisfactory returns to investors" (Taylor 1951, p 27).
There is no doubt that the advent of railroads exerted a tremendous effect on regional trade within New England, and with the rapidly developing Ohio Valley. Coastal traffic was relatively unaffected, but the movement of goods from the coast inland, in this region of few large waterways, was immediately stimulated. The Granite Railroad in Quincy, MA, opened in 1826, is sometimes described as the first American railroad, but it was really a gravity-powered tramway moving granite from the quarry to Boston Harbour. The first steam railroads in Massachusetts were the Boston & Lowell, the Boston & Providence and the Boston & Worcester, all incorporated in 1830. The Lowell railroad immediately put the Middlesex Canal out of business, while the Worcester did the same for the Blackstone Canal. Railroads and canals were only compatible in a few areas of the U.S., such as the Great Lakes, where the Erie Canal, opened in 1825, remained an important link between the lakes and New England. The value of goods shipped West via the Erie rose from $10,000,000 in 1836 to $94,000,000 in 1953 (Taylor 1951). Meanwhile, 268,000 barrels of flour where shipped back from the Midwest to New England in 1835, and 1,000,000 barrels by 1840. Until the 1840's, textiles and other products had to be sent either by the Canal or by ship to Philadelphia and Baltimore, and overland from there. After the Western Railroad to Albany opened in 1841, this coastwise trade was gradually taken over and disappeared by 1860. During this period, interregional trade between New England and points west indubitably experienced rapid growth.
For the purposes of this inquiry, however, the important question is, to what extent were commercial "markets," according to the definition given above, a feature of New England life before the advent of the railroads and the "transportation revolution," versus afterwards. There are fewer good figures for the overall extent of domestic than for foreign trade, but detailed local studies show that there was a steady flow of goods within the New England region, including bulky goods and animals (Rothenberg 1992, Garrison 1987, Baker and Izard 1991, Clark 1979, 1990). John Adams, in his Diary and Autobiography, describes setting out on a springtime journey in 1771 from his home in Braintree to visit the Connecticut valley. Stopping in Waltham for a mid-day meal, he is interrupted by a large herd of cattle being driven past the window on their way to the slaughterhouses in Brighton, MA. These were representatives of the flourishing and long-standing New England trade in fat cattle , a system in which cattle were bred in upland farm communities as far north as northern New Hampshire and Vermont, sold in the autumn to farmers in the fertile lowlands of the Connecticut River, fattened over winter on their considerable grain surpluses, and sold in the spring Boston markets (Garrison 1987). After the War of 1812, with its embargos and high tariffs, this trade became increasingly competitive, and many Massachusetts farmers turned instead to growing broom corn for making brooms, as well as other specialized cash crops such as hemp, broom corn and tobacco. Farmers with access to capital and particularly good lands were able to continue in competition with imports from the West, but by the 1840s, this was becoming increasingly difficult. In 1837, Northampton historian Sylvester Judd wondered "whether this gambling, speculating age for money will ever receive any permanent check" (quoted in Clark 1979).
Thus, there is evidence that involvement with an established market economy in New England can be traced to at least the last quarter of the 18th century, well before the advent of modern transportation methods. Since the loss of forest cover was already gaining momentum by 1800, it appears that forest loss must not have depended on efficient transportation systems, but was correlated with the growth in extent and importance of markets.
However, it is clear that the extent of market involvement and interregional trade has continued to increase steadily, even explosively, into the 20th century. Yet after the Civil War, forest cover began to return throughout New England, and has continued to increase until very recently. So, it is not the market economy per se that is correlated with forest loss, but some component of market economy. What happened in New England to produce the turn-around in forest loss?
Urbanization: Manufacturing and Trade
Urbanization, or the growth of cities, is often cited as an important factor in the recovery of the New England forest cover in the second half of the 19th century. But urbanization is a complex phenomenon with multiple interacting root causes. The urbanization process in New England was really two separate processes: cities grew, and the proportion of the population living in cities increased relative to the rural population. These processes need not always happen simultaneously, and they may result from different causes (Riefler 1979). According to one model in urban studies, cities may be seen as relatively homogeneous regions (in respect of demography, labor markets, communications, etc.) which grow as a result of external relationships. These might include inter-regional trade, intercity communications and competition. Another model, leaning on "central place theory," emphasizes the relationship between a city and its "hinterland," or the rural areas surrounding the city and contributing goods and resources to the urban population. This model points to the expansion of intra-regional trade and the economies of scale in the provision of urban services which become available with growth. Which set of relationships was more important in the growth of New England cities in the 19th century? And, since forest cover continued to decline until the middle of the century, then began to increase, while cities continued to grow throughout the century, can we distinguish different processes at work before and after the Civil War?
These questions can be restated: in the complex phenomenon known as urbanization, are some elements correlated with deforestation, others with afforestation? For example, does trade over large distances (both stimulating and resulting from improved communications networks) tend to concentrate activity into urban areas and reduce agricultural extensivity? Does the symbiotic relationship between a city and its hinterland, expressed in the heightened pace of trade within a region, exert a stronger influence on the urbanization process? Or does the increasing trend toward specialization and growth of manufacturing contribute more to urbanization? And which kinds of processes may be expected to correlate with reduced forest cover?
Riefler (1979) provides data which hint at an answer to these questions, although he analysed a data set including 103 urban areas covering the entire Northeast quadrant of the U. S as far south and west as Louisville, KY. Therefore, his published results are perhaps only suggestive as regards New England, which was without doubt an exceptional region within the U. S. as a whole. Nonparametric correlation tests of the growth rates of cities and their hinterlands between 1800-1830 show these two variables to be significantly related, indicating the importance of intra-regional trade to the growth of cities during this period. The symbiotic relationship between the urban core and its surrounding rural area stimulated the growth of each. In contrast, the growth of manufacturing appears to show no statistical correlation to the growth of cities before 1860, implying that although manufacturing was growing during this period, it was not spurring the growth of cities. As regards New England, this finding is consistent with the tremendous growth of small rural mill villages which sprang up across the rural landscape during the decades leading up to the Civil War (Kulik et al. 1982), giving a far-flung, pastoral character to the nascent industries of the region. It should be noted that, in the Riefler study, New England contributed half of the cities within the entire Northeastern region which were more specialized in manufacturing than was the U. S. as a whole (Riefler 1979). By 1860, therefore, New England was already a relatively specialized region. There was a group of New England cities, such as Lowell and New Bedford, which were clearly specialized in manufacturing by 1860. But even for this group, there was little relationship between population growth in the urban areas plus their hinterlands, to population growth in the urban cores themselves. This relationship may be taken as a proxy for the growth of urban manufacturing, since ***
This indicates again that, even in New England, manufacturing was not closely correlated with city growth during this period. Intra-regional trade - trade between cities and their hinterlands - still seems to have been the primary stimulus to urbanization during the first three decades of the 19th century, so that cities and their hinterlands grew in tandem.
Between 1830-1850 (the years which saw the most drastic reductions in forest cover) inter-regional trade between different cities takes over as the most important influence on city growth, undoubtedly as a result of the fast-developing rail network throughout the country. These were the decades that saw the first major agricultural specialization in New England: namely that for sheep. The trend began in the southern and middle New England states and moved north. The northern hill country was virtually turned into sheep pasturage by the 1840's, when Vermont had 5.75 head of sheep per inhabitant (in contrast to one per person in Virginia, Connecticut, and Pennsylvania) (Wilson 1967). "Men counted their flocks by the thousands, and as they grew more and more rich in money and sheep, they bought farm after farm adjoining their own and turned them into pasturage" (Thrasher, A New England Emigration, 1897, quoted in Wilson 1967). The wool from Vermont and New Hampshire was sent to mills in Massachusetts and southern New England, and stimulated trade both within the region and with other regions. It has been asserted, however, that the New England wool industry was built upon a protective tariff on imported wool (Wilson 1967). When this tariff was removed, the New England industry went into decline and quickly disappeared, to be replaced by the dairy industry (but see Taussig 1900 for a contrary evaluation of the wool industry).
After the Civil War, manufacturing and industrialization seem to assume primary importance in the growth of cities, while intra-regional trade - the trade between cities and their hinterlands - no longer appears to be a significant influence. Since the period after the War (1860-1900) saw an overall reduction in deforestation rates and a reestablishment of forest cover, it appears that an urbanization process emphasizing the growth in importance of urban manufacturing and industry was, in the case of New England, correlated with increasing forest cover. During this later period, trade explains very little of the variation between cities (Riefler 1979).
Although the Riefler study treats a wide range of cities in regions of very different character, it does, by examining the relative explanatory power of several variables in the urbanization process, reveal this process as amenable to analysis. The growth of U.S. cities during the 19th century, and the increased concentration of population within urban areas, was not a single process, but one stemming from several different causes. Each of these was a decisive influence at different times, and each was correlated with a different trend in forest cover.
This preliminary survey of several economic factors which have contributed to the U-shaped trend in forest cover in New England, reinforces the importance of the switch from an extensive agricultural economy to urban-centered manufacturing and industry. There are many reasons to believe that in some areas within tropical developing nations (as in New England during the 19th century), agriculture may already have reached a level of intensity which is threatening the health of the ecosystems in which agriculture is embedded. Ecologically, intensive agriculture may not be sustainable in many areas. There are suggestions that a trend toward the consolidation of agriculture into ecologically robust regions, and a movement of populations toward urban areas, may be an inevitable result of market forces. Clearly, such a trend will continue to bring with it many difficult problems in urban design. It may also have the effect of a return of forest cover in some areas. Costa Rica, for instance, has reached a level of 15-30% remaining unprotected forest cover, including remaining primary and increasing secondary forests (FAO 1997). This happens to be approximately the same number as the estimated forest cover in New England was at its lowest, about 1860. Does this number imply the end of an "agricultural frontier," retrenchment from extensive agriculture, and an ensuing movement in the society as a whole toward other kinds of involvement in the market economy? The possibility of defining economic indices to make such a prediction is intriguing.
ConNE Home Page | Terrestrial Group