Let me begin with a quick heads-up on my forthcoming book. It’s been somewhat delayed in the editing, but Covid-19 permitting it’s now slated for publication at the end of October. So please be sure to keep some cash in your piggy bank for the tail-end of the year…
One reason the book was delayed in the editing is because the initial draft grew a little unwieldy and I’ve had to spend time paring it down. There’s just so much to say about smallness, farming and the future! Some of my edits are destined to languish forever on the cutting room floor, perhaps rightly so, but there are a few sections I think deserve to see the light of day. So I’m aiming to publish them here on the blog as a kind of ‘best of the rest’ selection – or, to a put a more positive spin on it, as background reading that fills out in greater detail some of the book’s balder and briefer assertions and analyses.
This post is the first of these selections, lightly edited to fit into the blog format, involving some reflections on farm scale, yield and income. Right now, however, I’m in the thick of the final book edit, so please forgive me if my replies to any comments are more peremptory than normal.
oOo
For numerous reasons, I’ve long argued for a small farm future, where a large proportion of the population work as small-scale agricultural proprietors producing food both for themselves (a crucial point, as I’ll emphasize below) and for market sale. But it has to be said that historically the lot of the small proprietor often hasn’t been a happy one.
Optimal in theory, sub-optimal in practice – in his book Agricultural Revolution in England1, Mark Overton includes an interesting table that enables us to probe this issue. The table presents data from two pre-industrial English grain farms, which are hypothetical but presumably grounded in Professor Overton’s experience as a prominent agricultural historian – a large farm of 100 acres and a small farm of 10 acres:
| Farm Productivities | ||||||||||
| Column
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
|
| Row | Harvest ratio to normal | Acres | Yield (bushels/acre) | Gross output | Seed | On farm consumption | Net output | Price (d/bushel) | Total income | Income/labor unit |
| Large farm | ||||||||||
| 1 | 1.5 | 100 | 15 | 1500 | 250 | 50 | 1200 | 4 | 4800 | 4800 |
| 2 | 1.2 | 100 | 12 | 1200 | 250 | 50 | 900 | 6.6 | 5940 | 5940 |
| 3 | 1 | 100 | 10 | 1000 | 250 | 50 | 700 | 10 | 7000 | 7000 |
| 4 | 0.8 | 100 | 8 | 800 | 250 | 50 | 500 | 16.9 | 8450 | 8450 |
| 5 | 0.5 | 100 | 5 | 500 | 250 | 50 | 200 | 55.3 | 11060 | 11060 |
| Small farm | ||||||||||
| 6 | 1.5 | 10 | 15 | 150 | 25 | 50 | 75 | 4 | 300 | 300 |
| 7 | 1.2 | 10 | 12 | 120 | 25 | 50 | 45 | 6.6 | 297 | 297 |
| 8 | 1 | 10 | 10 | 100 | 25 | 50 | 25 | 10 | 250 | 250 |
| 9 | 0.8 | 10 | 8 | 80 | 25 | 50 | 5 | 16.9 | 84.5 | 84.5 |
| 10 | 0.5 | 10 | 5 | 50 | 25 | 50 | -25 | 55.3 | -1382.5 | -1382.5 |
In Rows 1-5 Overton gives some input/output figures for the large farm under various assumptions of good, normal and poor harvest. In an agricultural economy that’s largely self-sufficient in staples locally or nationally, demand for grain is price inelastic (ie. it stays fairly constant regardless of price, because everybody needs to eat). Because of this inelasticity, in times of dearth the price of grain shoots up disproportionately to the fall in output, meaning that the large farm gets more income in bad harvest years than good ones (compare Row 1, Column 9 with Row 5, Column 9).
The outcome isn’t so happy for the small farmer, shown in Rows 6-10. In the poorest harvest years, s/he produces less grain than s/he needs to consume (Row 10, Column 7), and has to buy in extra grain to eat at the inflated scarcity price, therefore making a net loss (Row 10, Column 9). Too many seasons like that and the farm goes under, forcing the farmer to find some other employment – if they can.
But Overton makes a key and rather hidden assumption. Before it sells any grain the farm household first has to meet its own need for sustenance, which is the same year by year for a given household size regardless of the harvest. Overton allows for this in Column 6, but he makes it the same for both the 100 and 10 acre farms. So the same number of farmworkers in both cases (and the same yields per acre) but on the large farm the same number of workers are applied to an area ten times the size of the small farm, and therefore have ten times the labour productivity.
That’s a reasonable (in fact, conservative) picture of what tends to happen when small-scale farmers with hand tools or draft teams confront large-scale ones with all the paraphernalia of modern fossil-fuelled industrial agriculture. The labour productivity of the latter is vastly greater, and since labour is a major cost driver, this pushes poor small-scale farmers out of staple crop production and into commodity crop or non-farm employment markets where they’re subject to wider market forces, usually to their disadvantage. This is why calls in the rich countries to improve labour-shedding technologies, increase yields and lower the price of food in order to ‘feed the world’ in fact are more likely to starve it.
But I’m not sure Overton’s labour productivity figures present a reasonable picture of preindustrial English agriculture, or situations generally where the large-scale farm has no technology or labour-productivity advantage. Suppose we recalculate Overton’s table assuming that each acre of farmland requires the same amount of human labour applied to it. It then looks like this:
| Farm productivities scaled to labor productivity | ||||||||||
| Column
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
|
| Row | Harvest ratio to normal | Acres | Yield (bushels/acre) | Gross output | Seed | On farm consumption | Net Output | Price (d/bushel) | Total income | Income/labor unit |
| 1 | 1.5 | 100 | 15 | 1500 | 250 | 500 | 750 | 4 | 3000 | 300 |
| 2 | 1.2 | 100 | 12 | 1200 | 250 | 500 | 450 | 6.6 | 2970 | 297 |
| 3 | 1 | 100 | 10 | 1000 | 250 | 500 | 250 | 10 | 2500 | 250 |
| 4 | 0.8 | 100 | 8 | 800 | 250 | 500 | 50 | 16.9 | 845 | 84.5 |
| 5 | 0.5 | 100 | 5 | 500 | 250 | 500 | -250 | 55.3 | -13825 | -1382.5 |
| Small farm | ||||||||||
| 6 | 1.5 | 10 | 15 | 150 | 25 | 50 | 75 | 4 | 300 | 300 |
| 7 | 1.2 | 10 | 12 | 120 | 25 | 50 | 45 | 6.6 | 297 | 297 |
| 8 | 1 | 10 | 10 | 100 | 25 | 50 | 25 | 10 | 250 | 250 |
| 9 | 0.8 | 10 | 8 | 80 | 25 | 50 | 5 | 16.9 | 84.5 | 84.5 |
| 10 | 0.5 | 10 | 5 | 50 | 25 | 50 | -25 | 55.3 | -1382.5 | -1382.5 |
By equalizing the labour productivities in Column 6, the advantage of the large farm has disappeared. Like the small farm, its income turns negative in the poorest harvest years, and its financial returns per unit of labour are exactly the same. In fact, in preindustrial England and in many other places historically the evidence suggests that, if anything, there are diseconomies of large scale in terms of output per acre, so the small farm may be advantageously placed.
There are four points I’d like to draw out from this little exercise by way of conclusion.
First, there isn’t some natural economic law that favours large over small economic units. Only in specific social and technical circumstances is this likely to be the case.
Second, one such case has been global agricultural mechanization over the past century or so. If like units of labour earn more or less the same per hour whatever they do, and if there are no diseconomies of increased scale in relation to labour-shedding agricultural mechanization, then small farms producing grain by hand or small machine will be disadvantaged relative to larger farms that are terraformed to the requirements of large machines employing equivalent labour. What seems much less clear to me is that this will continue to be true into the future. In the coming years we’re probably going to have to find low carbon employment for people in their multitudes. In this situation, labour productivity will probably be less important than carbon intensity and gainful employment – and the small farm may be better fitted to that end.
Third, a talking point in mainstream agricultural economics is that ‘free’ markets rather than household or community self-reliance are a better safeguard against hunger. Sometimes that can be true. The exorbitant prices for grain apparent in Column 8 as the yields in Column 3 decline might have been smoothed out with imports of cheaper grain from other parts of the world experiencing grain surpluses.
But there are several grounds for caution here. There’s the problem of dumping I mentioned above, with long-term negative effects on the local economy. There’s the issue that while cheaper grain imported from elsewhere may be welcome when harvests are poor, the exporters are usually price-seekers, not humanitarians. The cheaper grain may not be cheap enough to fully relieve distress – distress arising largely from local social arrangements which wider market dynamics aren’t geared to mitigating. Indeed, speculation on financialized global markets is a driver of food price increases, not a wider market solution to local distress. Finally, in the climate and water-challenged world that’s now upon us, it’s unwise to assume that cheap grain at steady prices will be readily available on the global market. The places that produce most of the grain surpluses are among the ones that are most climate and water challenged – and, as we’ve seen recently, when push comes to shove administrations prioritize production for domestic use, at least for those among their populations they wish to court.
Fourth and finally, inasmuch as farms of any scale might be economically threatened by poor harvests, it seems to me that Overton’s analysis points to two remedies. The first is that the farm household should place a high priority on producing for its own needs in a resilient, pessimal manner so that the net output figures in Column 7 are unlikely ever to turn negative, even in bad years, as a matter of ecological management. The second is that the farm household should aim to be part of a wider community and economy of farm households whose political guardians offer them support so that in bad years the figures in Column 7 don’t turn negative as a matter of political fiat. I think both of these remedies will be necessary in a sustainable and resilient small farm future.
Generally, the point I want to stress is that the bad outcome for the small farmer of the kind highlighted by Overton doesn’t arise from some intrinsic disadvantage of small scale, but from relative labor productivities in the existing globalized capitalist market – which isn’t the only way or necessarily the most sensible way of organizing food production and social wellbeing.
Reference
- Mark Overton. 1996. Agricultural Revolution in England: The Transformation of the Agrarian Economy 1500-1850. Cambridge University Press.

