Sunday, November 01, 2009

Why Malthusianism refuses to die

kitten in a basket by pixn8tr from flickr (CC-NC-SA)
It's astonishing how popular Malthusian catastrophism is, in spite of such spectacular evidence to the contrary. As disaster keeps failing to come, its details are changed, but the core prediction that we're all doomed doesn't budge. In this post I'd like to explore some evidence against Malthusian ideology - as with so much evidence to the contrary, it like creationism long ceased to be anything else than ideological view.

Classical Malthusianism

But first, let's say what I'm criticizing. Malthus' idea was that population growth is exponential, while food production growth is arithmetic, therefore we're all going to fall back into extreme poverty due to overpopulation, until we're poor enough to be dying in sufficient numbers to counteract that.

The second part is, and I'm not making this up - "therefore, we should starve and abuse people who lost their jobs in a recession to lower taxes on the rich". That's a forerunner of the Real Business Cycle theory if there ever was one. The predictable result of being tough on the poor was that the working poor were starved to the point of eating pig fertilizer, beaten, and sexually abused.

But that's not the worst of Malthus-inspired politicians - they decided to starve one million poor people in Ireland. During the alleged "famine", Ireland was a major exporter of food, with its food production even at lowest point estimated to be enough to sustain population twice as high as it had. And there was well known solution to such famines already tried in 1782–83 of temporarily stopping food exports, but those good Malthusians would rather have a bit of a genocide than to stand in ways of profits.

Anyway, I'm not here to criticize consequences of Malthus' ideology, just its factual basis. It's been 211 years since Malthus' infamous essay, how far behind are we in food production?

Oh wait, actually food production looks pretty exponential on this graph, while population growth looks rather linear. It's not even that - according to virtually all projections population growth is slower than linear, and world populations are due to stabilize fairly soon, here's Wikipedia:
Unfortunately I haven't found any useful food production statistics going all the way back to 18th century, but by now it should be very clear that these relationships do not hold.

Yet, Malthusianism refuses to die. The most popular mid 20th century's neo-Malthusian Paul Ehrlich famously wrote in 1968:
The battle to feed all of humanity is over. In the 1970s and 1980s hundreds of millions of people will starve to death in spite of any crash programs embarked upon now. At this late date nothing can prevent a substantial increase in the world death rate...
Or more specifically - "I have yet to meet anyone familiar with the situation who thinks India will be self-sufficient in food by 1971" and "India couldn't possibly feed two hundred million more people by 1980."

That was in the middle of yield increases so spectacular they caused major difficulties to even harvest it all (High yields led to a shortage of various utilities — labor to harvest the crops, bullock carts to haul it to the threshing floor, jute bags, trucks, rail cars, and grain storage facilities. Some local governments were forced to close school buildings temporarily to use them for grain storage. as Wikipedia says). India became net exporter of cereals by 1978.
You'd think that after two centuries of predictive failures people would have given up, but no. The latest neo-Malthusian fad just a few years ago was "peak oil". It sort of fizzled since, but in case you forgot, here's an awesome YouTube video describing peakoilnik ideology while dancing:



"There won't be much food anyway, as oil is used to make fertilizers and pesticides. Up to two thirds of the world population may not make it. Four billion people may not survive."

Because as they say, we're basically "eating oil" as the meme says. It doesn't bother them much that oil is used at no step of fertilizer production - it's all natural gas, and entire 3-5% of world natural gas (1-2% of world energy use) is spent on fertilizer production (not even as energy source, more as convenient hydrogen source easier than electrolysis of water), but alternatives based on coal, and hydroelectricity are well known and used depending on local availability. China already produces most of its fertilizer with coal. The first commercial hydroelectricity-based fertilizer plant was opened in 1905, over a century ago, and in case you're wondering use of electric plowing equipment dates to late 19th century.

Pesticide production is so ridiculously low, by some back of an envelope estimates takes about 0.06% of world's oil, and is trivially replaceable by coal if need be (I know of no serious estimates, as it's too small to bother anyone with a clue).

What a Malthusian world would look like

So why so many people believe in Malthusian-inspired catastrophism? I have a two-fold hypothesis. First, some people just enjoy beliefs that we're all going to die. It seems that virtually every disaster scenario, no matter how plausible or not - attracts countless believers.

But there are people smarter than that who still believe in Malthusianism. Here's the second part of my hypothesis. When people look at history, to see if it matches some theory, they just look at things that would be observable if theory was true, and see if there are present. They don't do proper Bayesian updating.

Here's a simple example - let's say you have a theory which says that clash of civilization causes warfare. Then the naive thing to do would be to look at history, note a few wars between civilizations, and take it as evidence for the theory. It would also be wrong, as far more numerous wars were fought within civilizations, and proper Bayesian thinking must look at both.

So let's take a look at evidence for Malthusian worldview against the null hypothesis that food production is just like any other commodity, and not any more connected to population than production of clothes, bricks, or paper.

If Malthusianism was true, food production per capita would be pretty much constant - because any increase in food production would result in increase in population, and decrease in food production would cause mass starvation, and adjust population accordingly. But if Malthusianism was false, food production per capita would also be pretty much constant, because what would be the point of growing more food if there was nobody to eat it? So there's no evidence either way.

If Malthusianism was true, everyone would be poor. Not only is this ridiculously false, now, historical estimates of life expectancies vary wildly from 18 to 60 for the same time period.

Malthusianism predicts almost all food production would be in whatever generates highest number of calories per hectare. Non-Malthusianism predicts food production would be much more varied - and it is so. Just the fact that we're keeping farm animals, which by their nature only destroy and not produce nutrients, proves Malthusianism wrong. There might be exceptions, but most grazing lands would produce more calories when cultivated, not to mention Malthusian ridiculousness of feeding animals human food as we do.

If Malthusianism was true, any sudden decrease in population, due to wars, famines and such, would result in rapid rebound, as agricultural capacity existed to support much larger populations, proven by its previous existence. If Malthusianism was false, population would have no particular reason to rebound. Here the evidence is very clear - there are numerous example of major population loses which were not recovered for centuries. Two best examples affect entire Europe, which used to have high populations at height of Roman empire, then fell to very low levels and didn't recover until 12-13th centuries, almost a millennium later. Second fall was due to Black Death, recovery from which wasn't complete even two centuries later.

Malthusianism predicts that it would be extremely unlikely for vast amount of spare arable land to exist - yet this was exactly the situation for centuries in early Medieval Europe, with vast tracks of unused lands.

Malthusianism predicts that famines would happen during times of higher population a lot, while almost never happening during time of low population. But even a brief look at Wikipedia's List of famines shows plenty of major famines both during population lows of early middle ages (5th century famines in Western Europe, 750's famine in Spain, 809 famine in Frankish Empire, 963-964 in Ireland, 1005 in England, 1016 all Europe, 1030-1032 France, 1066 England again, 1097 France again). There were supposedly 95 famines on British Isles alone during Middle Ages. In spite of very low population levels, we can see plenty of them, and we'd probably know about more if it wasn't for horrible Medieval record keeping. Notice that during all that time there was plenty of undeveloped arable land!

During second period of low populations after Black Death, there are major famines 1390 in England, 1481-1483 in France, 1504 in Spain, 1518 in Venice and so on. I'm not even going to mention that modern periods of European population explosion had fewer famines than any other time in history.

Non-Malthusian view correctly predicts that famines and population levels are completely uncorrelated. Famines are caused not by the mythical "overpopulation", but commonly by wars, government action (like in Ireland, and 1930s' Ukraine, notice how democratic governments have virtually spotless track record of no famines, in spite of plenty of agricultural disasters), and most of all - simply by major delay between increase in agricultural inputs and outputs. It doesn't matter that there's plenty of unused land around, developing it would take years, and plenty of labour, capital, seed, and farm animal input, before any food was grown on it - all of which being in short supply during famine anyway. By the time food was growing the famine would be long over. To give you some indications of time scales involved, settlers in new developments were freed from any taxation for up to 24 years. During that time net effect on food production would be negative.

Most of all, current overwhelming explosion of wealth is completely and utterly incompatible with Malthusianism. Even the second worst country in the world - Angola - has life expectancies far greater than world had 100 years ago (one tiny country which doesn't has 26.1% HIV infection rates, and is too poor to afford HIV drugs - hardly a Malthusian reason). In spite, or maybe even due to, population explonion, world's average life expectancy is 66.57 now. This is level not reached by even the richest countries before 1930, and by most developed countries well into 1940s an 1950s.

However hard I try, I cannot find any evidence for even very broadly constructed Malthusianism that cannot be explained as well or better by treating food as a normal commodity. So worry not - we're not doomed.

6 comments:

  1. Economic historian Gregory Clark — author of the excellent **A Farewell to Alms** would disagree with your assessment here — until 1800, Malthus absolutely was correct.

    Clark illustrates with charts and data that Malthus was right about most of human history —excepting the last 2 centuries of civilization history: living standards basically did not improve from the age of antiquity to the age of Louis XIV, because the little technology gains were swallowed up by population pressure.

    Not until the industrial revolution did the world break free of Malthus model.

    But what's to say that the last 2 centuries are not just an anomaly? Or is it just a given that that cord has been broken forever?

    ReplyDelete
  2. Naum: Have you read this post? There's a long list of arguments why Malthus was historically wrong.

    ReplyDelete
  3. Yes I read your post and most of your data cites are post industrial-revolution (after 1800) and pre-1800 references are all anecdotal compared to Clark's exhaustive data dump. Clark goes through English records predating 1800, also proffers a gauge of technology, per capita caloric intake, wages, etc.… Basically, before 1800, state of average person was not much different than times of antiquity.

    After 1800, indeed, Malthus theory crumbles.

    But for except the last 2 centuries, it's been illustrated by Clark and other economic historians to be close to spot on.

    And my question still stands: can we assume that technological progress that deprecated Malthus in the 19th and 20th centuries is a given for centuries forward?

    ReplyDelete
  4. Medieval Europe did not happen in the last 200 years. Massive demographic shifts during that time are all extremely well documented and there's nothing anecdotal about them. There's also wide and well documented variation in human lifespans, and other quality of life measures over historical times.

    Looking at caloric intake, wages etc. is completely ass-backwards. These would be the same with and without Malthusian hypothesis, which states that population is determined by capacity of food production. This is trivially and provably false. Food especially, as there's very little variation in demand per capita.

    People were mostly poor in most (but not all) historical eras due to the same reasons that Irish starved to death during potato blight time - those in power didn't care about them at all. There's no evidence whatsoever it had anything to do with agricultural capacity.

    ReplyDelete
  5. With all due respect, you have not even approached even denting Clark's (and other economic historians) case…

    If you haven't read "Farewell to Alms" you should, it's as relevant or greater than Adam Smith "Wealth of Nations", even if you disagree with the author's research and/or hypothesis…

    ReplyDelete
  6. Quickshot12:44

    Nahhh, I think he has a point here, why in hell would have famines when your population declined and didn't recover right away? Well I guess the farms could be abandoned.... but still.... just those few data points are rather suspicious.

    More seriously there is a reason to think it doesn't apply to the modern era at all, namely that once humans gain a small modicum of wealth, +-$2000 is about enough already, they naturally drop reproduction below self sustaining levels. Of course there are reasons why it happens at about that level. And it isn't the actual money itself, but it's an easy way to keep track.

    But basically what this means is, is that now that the planet is getting more wealthy the population is slowing down in growth and by 2050 we might be already flat level, which means you have a population level completely independent of food supply. That level isn't even catastrophic, because it's easy to see with our massive meat uptake that there is a fair amount of reserve food supply still available and we could always use more expensive investments like greenhouses to further increase returns as well.

    So regardless if it was true in the 18th century and past, there is absolutely no reason to believe that this trend is going to continue in the future, as it seems like we've entered a different population control regime now.

    ReplyDelete