Malthus redux? 3. A little historical background

In the first post of this mini-series it was pointed out that, after more than a century of decline, relative food prices seem to have started a secular increase again. In the second it was pointed out that poor households all over the world are already suffering because of this. In this post it will be argued that, in a historical perspective, ‘Malthiusian strains’ were overcome only quite late in the process of ‘modern economic growth’. A clear and sustained increase of ‘total’ GDP production was not enough for what Robert Fogel called ‘The conquest of high mortality and hunger in Europe and America‘ (let alone the rest of the world). And it’s still not enough – surely not when food prices increase. A crucial ingredient of genuine progress seems to have been a shift from market to home production by mothers, enabled by higher real wages for men. There might have been very, very good reasons why the ideal of the non-working mother was so strong between 1880 and 1960. Which points our attention to the crucial role still played by households when it comes to genuine prosperity and progress.

Serendipity: just the other day Brad deLong published an extensive blogpost about the same theme, which is fully consistent with the graph below and provides valuable background information. One might also consult this recent blogpost from Tyler Cowen, about among other things real english wages which only started to rise several generations after the onset of the industrial revolution. Aside: it is important to note that the three most important ‘personal’ Anglosaxon economic blogs (the ‘Conscience of a Liberal’ by Paul Krugman, ‘Marginal Revolution’ by Tyler Cowen and Alex Tabarrok and the Brad deLong blog) are all written by people who know their share of economic history and write their blogs to change the ‘unquestioned’ concepts of economists.

Anyway. The graph shows a series on two centuries of per capita availability of food in the Netherlands, based upon this publication as well as on FAO data for 1961-2005. For a very important non-western series see this series for Indonesia from Pierre van der Eng (oops, happened to be my academic roommate, about twenty-five years before we published these series). For data even further back in time also this publication from Riemke Westerholt about food consumption in a Groningen orphanage, 1631-1831 or this one from Anne McCant for food consumption of orphans in Amsterdam).

What does the graph show (to prevent misunderstandings: protein is an index, actual per capita availability in 1807 was 106 grams)? Background: around the beginning of the nineteenth century the Netherlands (basically a drained swamp) still were a rich country despite about a century of de-industrialisation, among other things thanks to productive, specialized agriculture, an efficient transport sector (the drainage canals…) and extensive trade in grain and colonial products based upon international trade and international division of labour, comparative advantages and a fair amount of outright exploitation, slavery and violence, not least in the Baltics where much of the grain imported by the Netherlands originated.

During the first seventy years of the nineteenth century the country witnessed, according to Van Riel and Van Zanden, a considerable increase of GDP per capita. At the same time, the population increased from about 2 million to about 3 million, an unprecedented increase. Despite per capita economic growth and because of the increase of the population it became increasingly difficult to offset harvest failures (see point a in the graph) by the traditional Dutch response: increasing imports. Especially when (b) the potato blight of 1845-1848 was aggravated by harvests failures of rye (1851, 1854) and the Crimean war (1853-1856) which led to a decline of grain exports from this area, the situation was dire. Among other things because population increase had led to a sharp decline of the amount of cows per capita (a clear Malthusian development). This led to increasing problems: for one thing the average height of conscripts (a sensitive indicator of overall health and stamina) declined considerably. Only after the end of the civil war in the USA (1865), when (point c) USA exports of grain to Europe rapidly increased (which showed as lower relative food prices and therewith of real wages) an improvement of the food situation (and the height of conscripts) can be witnessed (watershed 1).  Important:  Dutch agriculture was, after 1850 able to provide a more or less stable amount of calories per capita, the increase of the availability of food was enabled by imports (see also points e1 and e2, when WW I and WW II caused an almost total disruption of imports). Consumption of protein however still is ‘only’ about as high as 200 years ago (the high level of protein consumption at the beginning of the century was caused by a very favourable cow/people ratio, which led to a whopping 36% of total calories provided by butter, cheese, milk and buttermilk (yes, I quintuple checked that: a feed balance, data on grass growth, the orphanage data etcetera)!

After 1895 food consumption per capita stagnated (d) and, surprisingly, the high correlation between availability of food and average height suddenly breaks down too (watershed 2) and  the height (and health) of the Dutch population continued to increase for generation after generation. This was probably caused by (here I depart a little from Brad deLong, who does to my taste not stress improvements in the ‘household production function’ enough), first, improvements in the domestic situation enabled by higher wages for bread winners which enabled more mothers to stay at home (think: more breast-feeding which led to lasting improvements in health and stamina of babies). The city of Tilburg which experienced a quite late phase of low wage industrialisation is the çounterfactual’. This development led, in a period when infant mortality declined all over the rest of the Netherlands, to decades of rising infant mortality as mothers had to work in the factories too and infants were left at home, in the summer heat with a bottle of spoiled milk (at best). Second, in a later phase, improvements in public infrastructure (sewer systems, health systems, education) played their part. See about the importance of ‘household time’ as opposed to time spent on wage labour also this study, ‘why time deficits matter’, by Zacharias, Antonopoulos and Masterson.

After WW II things get pretty boring from a food consumption history point of view. At least in the rich world. But not for poor households, all over the world, which often saw large improvements. Until recently – as relative food prices rise again. The main take aways (surely not new, but worth stressing again, and again) of this all:

(1) Economic growth and food security are not by definition the same thing, do not count on ‘trickling down’ when it comes to improving the food situation – surely not when food prices increase.
(2) Households are crucial. Patterns of economic growth which lead to ever larger amounts of paid labour by members of poor households can, even when money incomes rise, be detrimental to the health and stamina of the members of these households, especially of the most vulnerable members. Look at this paper from Harris, Fogel, Floud and Sul Hong which argues that health and stamina of the members of households are not only results but also crucial elements of the household production function – as well as critical determinants of modern economic development. And of course the main conclusion of the Fitoussi-Sen-Stiglitz committee on economic statistics was that economists have to pay more attention to households.

p.s. – I’m also slightly more positive than Brad deLong about post medieval European technological development than Brad deLong. Think of glasses, musical instruments, improvements in dykes and drainage technologies, the introduction of the potato and corn in Europe, better ships, much faster postal services, the printing press, clocks, the use of gun powder in copper mining, increased use of peat which, for one thing, enabled farmers to use manure for fertilizing instead of cooking, the increasingly widespread use of the spinning wheel, at least some better houses and whatever.