Taxation and the stagnation of cotton exports in Brazil, 1800 – 1860

by Thales Zamberlan Pereira (Getúlio Vargas Foundation, São Paulo School of Economics)

Port of Pernambuco. Emil Bauch, 1852. Brasiliana Iconográfica.

Brazil supplied 40 per cent of cotton imports in Liverpool during the last decade of the eighteenth century (Krichtal 2013). By the first half of the nineteenth century, however, cotton exports stagnated, and Brazil became the only major international cotton producer that decreased its exports to European countries. The reason for decline in production, despite increasing international demand during the 19th century, is not generally agreed on. Scholars have attributed the decline to high transport costs, competition from sugar and coffee plantations for slaves, Dutch disease from the increase in coffee exports, among others (Leff 1972; Stein 1979; Canabrava 2011). There is disagreement in part because previous research largely relies on data after 1850, which was after the decline of cotton plantations in Brazil.

In a new paper, I argue that cotton profitability was restricted by the fiscal policy implemented by the Portuguese (and, later, Brazilian) government after 1808. To make this argue I first show new patterns on the timing of the decline of Brazilian cotton. Specifically, using new data of cotton productivity for the 1800-1860 period, this research shows that Brazil’s stagnation began in the first decades of the nineteenth century.

The decline therefore cannot be explained by a number of factors. It took place before the United States managed to increase its productivity in cotton production and became the world export leader (Olmstead and Rhode 2008). Cotton regions in Brazil did not have a labour supply problem nor suffered from a Dutch disease phenomenon during the early nineteenth century (Pereira 2018). The new evidence also suggests that external factors, such as declining international prices or maritime transport costs, were not responsible for the stagnation of cotton exports in Brazil. As any other commodity at the time, falls in international prices would have to be offset by increases in productivity. In fact, Figure 1 shows that Brazilian cotton prices were competitive in Liverpool. From the staples presented in the Figure, the standard cotton from the provinces of Pernambuco and Maranhão had higher quality than from New Orleans and Georgia (and, hence, achieved higher prices), but “Maranhão saw-ginned”, which achieved similar prices, used the same seeds as the ones in US plantations.

Figure 1: Cotton prices in Liverpool 1825 – 1850
Source: Liverpool Mercury and The Times newspapers

So, what caused the stagnation of cotton exports in Brazil? I argue that the fiscal policy implemented by the Portuguese government after 1808 restricted cotton profitability. High export taxes, whose funds were transferred to Rio de Janeiro, explain the ‘profitability paradox’ that British consuls in Brazil reported at the time. They remarked that even in periods with high prices and foreign demand, Brazilian planters had limited profitability. Favourable market conditions after the Napoleonic wars allowed production in Brazil to continue growing at least until the early 1830s.

Figure 2 shows that when international prices started to decline after 1835, cotton was no longer a profitable crop in many Brazilian regions. This was especially pronounced for regions where plantations were far from the coast, which had to pay higher transport costs in addition to the export tax. To support that the tax burden decreased profitability, I calculate an “optimal tax rate”, which maximized government revenues, and the “effective tax rate”, which was the amount that exporters paid. Figure 2 illustrates that, while the tax rate by law was low, the effective tax rate for cotton producers was significantly greater than the optimal tax rate after 1835.

Figure 2 – Rate of cotton export tariffs, 1809-1850.

Facing lower prices, cotton producers in Brazil could have shifted production to varieties of cotton produced in the United States, which had higher productivity and were in increasing demand in British markets. As presented in Figure 1, some regions in Brazil tried to follow this route (with saw-ginned cotton in Maranhão), but this type of production was not profitable with an export tax that reached 20 percent. Brazil, therefore, was stuck in the market for long-staple cotton, for which demand remained relatively stable during the nineteenth century. Regions that could not produce long-staple cotton practically abandoned production.

Not only do the results provide insight to the cotton decline, but the paper contributes to a better understanding of the roots of regional inequality in Brazil and the political economy of taxation. Cotton production before 1850 was concentrated in the northeast region, which continues to lag in economic conditions to this day. As I argue in the paper, the export taxes implemented after 1808 largely targeted commodities from the northeast. Production of commodities from southeast regions, such as coffee, paid lower tax rates. Parliamentary debates at the time show cotton producers in the Northeast did demand tax reform. Their demands, however, were not met quickly enough to prevent Brazilian cotton plantations from being priced-out from the international market.

To contact the author:


Canabrava, Alice P. 2011. O Desenvolvimento Da Cultura Do Algodão Na Província de São Paulo, 1861-1875. São Paulo: EDUSP.

Krichtal, Alexey. 2013. “Liverpool and the Raw Cotton Trade: A Study of the Port and Its Merchant Community, 1770-1815.” Victoria University of Wellington.

Leff, Nathaniel H. 1972. “Economic Development and Regional Inequality: Origins of the Brazilian Case.” The Quarterly Journal of Economics 86 (2): 243–62.

Olmstead, Alan L., and Paul W. Rhode. 2008. “Biological Innovation and Productivity Growth in the Antebellum Cotton Economy.” The Journal of Economic History 68 (04): 1123–1171.

Pereira, Thales A. Zamberlan. 2018. “Poor Man’s Crop? Slavery in Cotton Regions in Brazil (1800-1850).” Estudos Econômicos (São Paulo) 48 (4).

Stein, Stanley J. 1979. Origens e evolução da indústria têxtil no Brasil: 1850-1950. Rio de Janeiro: Editora Campus.

Overcoming the Egyptian cotton crisis in the interwar period: the role of irrigation, drainage, new seeds, and access to credit

By Ulas Karakoc (Tobb Etü and Humboldt University) and Laura Panza (University of Melbourne)

The full paper from this blog post was published on The Economic History Review and is currently available on Early View at this link

By 1914, the Egyptian economy confronted a unique conundrum: its large agricultural sector was negatively hit by declining yields in cotton production, the main driver of the economy. Egypt was a textbook case of export-led development, because  cotton production and exports had dominated the country’s economy.  The decline in cotton yields, which came to be regarded as a “cotton crisis”, was coupled with two other constraints: land scarcity and high population density. Nonetheless, despite unfavourable price shocks, Egyptian agriculture was able to overcome this crisis in the interwar period.  The output stagnation between  1900 and the 1920s contrasts with the following recovery (Figure 1).

Figure 1. Egyptian raw cotton output, acreage under raw cotton and raw cotton yield, c.1890s-1940. Source: Annuaire Statistique (various issues)

Previous research documented that during the crisis the decline in yields was caused by expanded irrigation without sufficient drainage, which led to a higher water table and made cotton more prone to pest attacks (Radwan, 1974; Owen, 1968; Richards, 1982).  This problem was addressed when the government  introduced an extensive public works programme directed to drainage and irrigation.   Simultaneously, Egypt’s farmers changed their cotton cultivation from the long staple and low- yielding Sakellaridis to the medium-short staple and high yielding Achmouni.  This change reflected income maximizing preferences (Goldberg 2004 and 2006). Another important feature of the Egyptian economy between the 1920s and 1940s was the expansion of credit facilities to farmers. Cooperatives, and the Crèdit Agricole (1931) were established  to facilitate small landowners’ access to inputs and  small loans (Issawi, 1954, Eshag and Kamal, 1967). These credit institutions coexisted with a number of mortgage banks, among which the Credit Foncièr was the largest, servicing predominantly large owners. Figure 2 illustrates the average annual real value of Credit Foncièr land mortgages in 1,000 Egyptian pounds (1926-1939).

Figure 2.  Average annual real value of Credit Foncièr land mortgages in 1,000 Egyptian pounds (1926-1939). Source: Annuaire Statistique (various issues)

Our work investigates the extent to which these factors contributed to the recovery of the cotton sector. Specifically: to what extent can intra-cotton shifts explain changes in total output? How did the increase in public works boost production? And, what role did differential access to credit play? To answer these questions, we construct a new dataset by exploiting official statistics (Annuaire Statistique de l’Egypte) covering 11 provinces and 17 years between 1923 and 1939.

We find  that access to both finance and improved seeds significantly increased cotton output,  and the declining price premium of Sakellaridis led to a large scale switch to Achmouni.  By putting farmers’ choices and agency centre stage in our analysis, our study shows that cultivators’ response to market changes was fundamental to the recovery of the cotton sector. Access to credit was also a strong determinant of cotton output, and productivity-enhancing innovations in agriculture (Glaeser, 2010).

Surprisingly, perhaps,  our results show that the expansion of irrigation and drainage did not have a direct effect on output (in the same or following year). However, we cannot completely rule out the role played by improved irrigation infrastructure for two reasons: first, we do not observe investments in private drains, and thus we cannot empirically assess the potential complementarities between private and public drainage. Second, we find some evidence pointing to the cumulative effect of drainage pipes, two and three years after installation.

We also find that the structure of land ownership, specifically the presence of large landowners, contributed to output recovery. Thus, despite the attempted institutional innovations aimed at giving small farmers better access to credit, large landowners benefitted disproportionally from credit availability. This observation accords  with Egypt’s extreme inequality of land holdings.

To contact the authors:

Ulas Karakoc,

Laura Panza,


Eshag, E., and Kamal, M.A.,  “A Note on the Reform of the Rural Credit System in U.A.R (Egypt).” Bulletin of the Oxford University Institute of Economics & Statistics 29, no. 2 (1967): 95–107.

Glaeser, B. The Green Revolution Revisited: Critique and Alternatives. Taylor & Francis, 2010.

Goldberg, E. “Historiography of Crisis in the Egyptian Political Economy.” In Middle Eastern Historiographies: Narrating the Twentieth Century, edited by I. Gershoni, Amy Singer, and Hakan Erdem, 183–207. University of Washington Press, 2006.

———. Trade, Reputation and Child Labour in the Twentieth-Century Egypt. Palgrave Macmillan, 2004.

Issawi, C. Egypt at Mid-Century. Oxford University Press, 1954.

Owen, R. “Agricultural Production in Historical Perspective: A Case Study of the Period 1890-1939.” In Egypt Since the Revolution, edited by P. Vatikiotis, 40–65, 1968.

Radwan, S. Capital Formation in Egyptian Industry and Agriculture, 1882-1967. Ithaca Press, 1974.

Richards, A. Egypt’s Agricultural Development, 1800-1980: Technical and Social Change. Westview Press, 1982.