A briefing document prepared for the Royal Society and Association of British Science Writers by Mike Holderness.

On 4 and 5 December 1996, the Royal Society held a scientific meeting entitled "Land Resources: on the edge of the Malthusian precipice?" This document was prepared afterwards to summarize key issues raised by the speakers and to provide a list of helpful contacts for future reference.

March 1997

There are two ways of looking at the Malthusian precipice: as an edge over which we can rush like the Gadarene Swine ¹ into the abyss; or as the heights which we are climbing, rather successfully so far.

From the introduction by the Chair ²

SUMMARY

How hungry will our children be?

How many people will there be in the world by, say, 2050?

How much food will they be able to grow, for how much longer after that and at what cost?

In 1798 the English clergyman Thomas Malthus published his Essay on the Principle of Population ²⁰. He noted that population rises geometrically. That is, if 10 people occupy a tiny island and each generation has twice as many members as the last, succeeding generations will have 20, 40, 80, 160, 320... members. He suggested that food production can only rise arithmetically: each generation may have the capacity to feed 100, 120, 140, 160, 180, 200... people. The fifth generation of our islanders reach the "Malthusian Precipice".

"The power of population is so superior to the power in the earth to produce subsistence for man," Malthus concluded, "that premature death must in some shape or other visit the human race."

Almost all who study population growth are now convinced that Malthus was wrong. In 1998 the earth's population will pass 6 billion¹⁸. In 2030 there will be 8 or 9 billion.

But the world's population will stabilise - well before we find out, for example, whether there is a serious global warming effect, still less what it is. Our maximum numbers are likely to be 10 billion, or 12 billion, or a few more.

The Big Question then is: can that many people feed themselves? More difficult: can they do it without destroying the soil for the generations to come, or destroying other species? Will there be any wild places left?

0. Introduction: the problem of forecasting

In order to answer the Big Question, we need to know:

• How many people there will be (if food supply does not restrict the numbere)
• How much food they will be able to grow; this in turn depends on many things, including:
  • How much land is available for food production
  • How much water can be delivered to that land
  • The type and condition of the soil on that land
  • How much nutrient - "plant food" - is available
  • How efficiently crops will be able to use those nutrients, and what possible future crop varieties are likely to be able to do

The answers to such questions depend on each other, and on yet other questions. "How much land is available" is a question of politics and economics as much as of geography. "How much nutrient is available" depends on the state of the soil and on the supply of water and on the capacity of possible future crop varieties to extract nourishment.

This briefing is divided into nine sections for easy reference. The real world clearly is not that neat: everything interacts. There is not space to list all the connections - please read the entire briefing!

Malthus' projections, as an example, were wrong because he reckoned without three things.

One is that over the last half century, through plant breeding, irrigation and use of fertilisers and pesticides - the so-called "green revolution" - humanity has managed a geometric increase in food production.

Another, which has brought the Malthusian Precipice closer to the present than he imagined, is the increase in life expectancy and reduction in infant mortality.

The third, which offers hope for our species, is the so-called "demographic transition". Crudely, when parents no longer need large numbers of children to work the land, they reduce the number of children they have. (There may be intense arguments about why demographic transitions happen; but there is no dispute that they do happen.)

It is not surprising, then, that forecasts can be spectacularly wrong.

(Though it was not mentioned in the meeting, the author was powerfully reminded that Joel Cohen, in his book How Many People Can the Earth Support? ²¹, reproduces a graph by Kenneth Blaxter. This shows the Registrar General's forecasts for the population of the UK in the year 2001, plotted against the year of the forecast. If you draw one smooth curve through the points, you may predict that in 2001 the Registrar will believe the UK population to be a large negative number. A large positive number of negative and hung-over people is more likely.)

We simply do not have the information to answer many of the questions accurately. Dr Webster ¹¹ pointed out that it is not good enough to look at soil type and quality on a national level; we would need a field-by-field survey of the whole world to produce accurate answers.

What forecasts can do is set upper and lower limits, provided nothing unexpected happens. It is not possible (barring meteorite strikes and the like) that the world's population in 2050 will be less than 6 billion; neither is it possible (barring a bizarre fertility-enhancing epidemic) for it to be much over 12 billion.

1. Population growth

Any mention of population invokes a horde of political demons: some Westerners' fears of teeming Third World hordes versus burning resentment of people for whom eating too much is a problem; the need for women to have choices over having children versus allegations of coerced abortion and anathemas on contraception... and so on.

Demographers - the people who predict future population sizes - now find themselves above these frays.

Dr Heilig³ reported that the lowest estimates for the world's population in 2050 are 7 billion, the expected number is about 9 billion, and the upper range 11 billion. The opinion that the lower range is likely is "not shared by many" - which is polite science-speak for "off the wall".

Discussion of population growth often - perhaps because of Malthus' continuing influence - focuses on rates of growth. Percentages do not eat. People do.

The number of additional people on the earth each year is already falling - Dr Heilig gives it as 85 million in the early 1990s, and 80 million a year now. The total population will probably level out at 12 billion or so.

This will happen regardless of anything that politicians or pundits do or say. Even wars and famines will have relatively little impact. The numbers are determined by the sheer number of children and teenagers alive now.

(In other words, the strongest constraint on the number of babies born is the number of mothers: and most of those who will be mothers in the next two decades have already been born. Medical advances which prolong the old age of the rich have a negligible impact. Advances which increase the numbers of children who survive to become parents have and have had a huge impact.)

The levelling-off of world population will, participants believe, occur through what are called "demographic transitions". Europe began such a transition when it was no longer necessary for parents to have large numbers of children to work the land. A vital factor in achieving a transition in the 20th century is educating girls beyond the 6th grade. South-East Asia, for example, seems to be in transition now - earlier than expected by some, and invalidating some of the more extreme population projections of the 1970s.

2. Land

Dr G. Fischer³ added that most of the increase in world population will happen between now and 2030. We have extraordinarily little time to develop the land sustainably.

How much land is there? Taking a broad-brush approach, Dr Fischer concludes that there are 1.6 billion hectares (Gha ¹⁹) of land which have some potential for cultivation and are not currently used. Two-thirds of that is forest and wetland - the Amazon Basin for example; 550 Mha have, "more or less", the ability to support cultivation.

But this potential new land is not in the same place as the people are: 45% of it is in Central and South America and only 8% in Asia. By 2050 the population of Asia will have passed the current population of the world.

The great bulk of the extra food required, therefore, will have to come from greater yields on already-farmed land. But in some areas - notably parts of Africa - productivity is declining as soil degrades or erodes.

3. Who decides? Economics for small farmers

In the areas with the greatest population increases, how food is grown is decided almost entirely by small farmers. They rarely have access to capital; loans, if available, are at extortionate rates.

Schemes for sustainable farming which require any interruption of crop supply or income are therefore futile. Someone who owns only clothes, cooking pots and seed-corn is not going to plant nut trees and then starve for ten years waiting for a crop.

Professor Gregory¹³ says that farmers in Australia, too, forcefully point out that what they're most interested in is staying in business in the bad years; if they didn't do that, they wouldn't be able to take advantage of the good years.

The very poorest people usually farm the most marginal land - that which is most vulnerable to erosion and other forms of degradation. Dr Barbier⁴ points out that for them it is often rational, in economists' terms, to farm in ways which maximise short-term returns.

Similarly, in places such as the Amazon where unfarmed land exists, poverty is a major driver of deforestation. In some cases - Dr Barbier mentioned ranching in South America and palm oil plantations in SE Asia - government policies on pricing or subsidised credit encourage deforestation, which encourages land erosion.

Where security of land-tenure is poor, whoever gets to unfarmed land first has effective control. People farming already-cultivated land have no incentive at all to do things like tree-planting (see Section 4.3) if they're likely to lose the use of the land at short notice. In some areas, almost all food production is by women - but any trees planted would belong to their men, or to their landlords.

Effective measures to conserve soil and forest must be aimed directly at small farmers on marginal land. Getting affordable credit to them would help a lot. Spreading information on sustainable farming methods which increase productivity - when many farmers are knowledgeable, but as "educated illiterates" - is vital. Time is short.

4. Land and crop productivity

How much food you can grow on a hectare of land depends fundamentally on (in no particular order):

• The amount of sunshine and length of the growing season - which we may be affecting on the longer timescale of global warming, but cannot alter at will;
• The amounts of nutrients available, from the soil and from fertilisers, and the absence of plant poisons like excess salt in the soil;
• The amount of water available to plants, from rain or from irrigation;
and
• The performance characteristics of the crop varieties being grown.

A fundamental principle is that a given plant will grow until it runs out of one vital resource. (Only if all resources are abundant do the plant's genetic limits come significantly into effect.)

If there is plenty of water but the soil is poor, nutrients are said to be "limiting". The same goes for individual nutrients: if there is plenty of water and nitrogen but little phosphorus, phosphorus is limiting. If there's plenty of everything else, elements which are required in tiny quantities, like zinc, may be limiting.

Water supply affects the availability of nutrients - how much the plants can actually get at. The condition of the soil affects both. Crudely, if the soil is baked to concrete, it may be chock-full of nutrients but they will be locked up, and water will run straight off.

Soil, water supply and yields have traditionally been studied by three very different scientific disciplines: soil scientists, hydrologists and plant physiologists too rarely, some said, follow each others' work. But, as Dr Wallace⁹ said in discussion, "The plants are not aware that two [sic] disciplines are arguing about them."

4.1 Crop efficiency

From the Division of Plant Industry in Canberra, Dr Evans ⁵ described the enormous increases in crop yields achieved by plant breeding (alongside soil and plant management). Much of this is due to short-stemmed or "dwarf" rice and wheat varieties, which can utilise large quantities of nitrogen fertiliser. Traditional grains, presented with that much food, get top-heavy, fall over and die. They also use more of the nutrient for stems and leaves than do the dwarves.

From time to time someone will notice a falling-off in yield increases, and conclude that the End is Nigh. Graphs of many processes of change start out with an upwards curve which could be the start of an "exponential" curve, representing a continuous percentage increase year on year, as below:

Few processes continue to follow this form of growth for ever. The streets of London are, in the hackneyed example, not in fact shoulder-deep in horse manure, as they would have been if an exponential increase in horse-drawn taxi traffic had continued indefinitely... Many processes of change instead follow an S-shaped or "sigmoid" curve ("sigma" is Greek for "S"):

Note that the early parts of the curves - the first 40 years in these arbitrary examples - are very similar. It is thus extremely difficult to tell when a process is makingthe transition. It is conversely very easy to "cry wolf" in either sense: "population is increasing exponentially!" or "food production is trailing off!" Dr Evans calls this latter "sigmoid fraud".

Further breeding can produce even more extreme dwarf varieties. A number of ways in which plants might use sunlight more efficiently are worth exploring, though the fact that we start from the results of more than 5000 years of selective breeding makes further staggering increases unlikely. Possibly the biggest opportunity for increasing grain yields is in producing varieties which are more precisely adapted to local conditions.

Dr Sivakumar⁶ presented a broad-brush approach to defining such conditions - the Agro-Ecological Zonation of the United Nations Food and Agriculture Organisation (FAO).

4.2 Water

It is, however, no good breeding efficient plants if they die of thirst. Dr Falkenmark⁸ presented what she called "back-of-an-envelope calculations" on the world's effective supply of fresh water.

Each person needs a minimum of 900 tons of fresh water a year, almost all of it for growing food. It is extremely expensive to capture and use more than 20% of the water which falls as rain.

On this basis, North Africa is running out of water about now. Northern Egypt already achieves more than 100% utilisation of water from the Nile, by re-cycling. And 26% of the water now used by Egypt is "virtual" - it is used in fields in other countries, from which Egypt imports food.

The uncertainties in Dr Falkenmark's projections of world water needs in 2025 are as large as the amount of water used today. But she concludes that very probably fresh-water supply will be a serious problem for more than half the world's people by 2050.

Dr Jim Wallace of the Centre for Ecology & Hydrology ⁹, by contrast, points out that a million tons of fresh water exists on Earth for each person. (A lot of it, though, is in inconvenient places like Siberia.)

Many things can be done to increase how efficiently water is used in crop fields. Radical changes in crops are unlikely to arrive in time, even with genetic engineering. Almost all the other measures are very low-tech and concerned with keeping the water which falls as rain where the plants can reach it.

"Mulching" with dead plant matter decreases evaporation from the soil. Terracing reduces run-off. Growing plants like clover underneath grain crops - often called a "green mulch" - also reduces run-off. Green mulches of, for example, clover and bean-family plants, "fix" nitrogen from the air - convert it to a form which plants can use - and thus fertilise the soil at the same time.

4.3 Soil conservation and nutrient restoration

Experiments in Kenya show that planting rows of trees, with "alleys" of grain or beans between, saves water equivalent to 70mm of rain a year.

The goal of this "agroforestry" was described by Dr Sanchez ¹⁰ as "successfully manag[ing] competition for light, water and nutrients between trees and other crops." In some places farmers can grow trees which produce valuable crops. Everywhere, they will provide fuel.

Some such conservation measures are urgent. Dr Sanchez estimated that Sub-Saharan Africa is losing 9.3 million tons (Mt) of crop nutrients a year through soil erosion and degradation. The region imports 1.7 Mt a year of fertiliser, so at most one-fifth of the annual loss is made up.

Dr Sanchez also described work on restoring soil fertility. Professor Syers¹⁵ predicts that investment in such restoration should have a rapid pay-back.

Where soil lacks nitrogen, both Dr Sanchez & Professor Syers propose mostly organic means to replenish it. (Nitrogen fertiliser is beyond the means of many African farmers.) Trees of the bean (legume) family can convert or "fix" nitrogen from the air into forms which plants can use - at a rate of 50 to 150 kg/ha/year. Recent work has also shown that their deep roots can get at nitrogen which cereal roots cannot reach. In some areas, however, this deep nitrogen may have been washed into the ground and wasted in the past; no-one knows whether it is a sustainable source.

Many farmers in Africa cannot afford manufactured fertiliser because of the effects of "structural readjustment" policies. Thousands are growing Tithonia shrubs, known as Mexican Sunflowers, in their hedges and using the foliage to mulch their fields. Apart from the other benefits of mulching, this allows the farmers to restore phosphorus-deficient soils using ground-up phosphate-bearing rock. This source of phosphate, Dr Sanchez says, is plentiful in Africa - enough for a century or more. Rock phosphate as it comes out of the ground is useful to crops only when it is applied with a mulch. In the industrialised countries it is chemically converted to super-phosphate form - an energy-intensive and expensive process. Replenishing phosphorus can increase annual maize yields from 1 t/ha to 4 t/ha.

Professor Vlek¹² and Dr Kuehne reinforced the idea that the key to soil fertility is a mixture of fertiliser and organic matter. Though mulches and agroforestry sound very "organic", not one speaker was opposed to inorganic fertilisers or to pesticides and several stated that without them we would starve.

Professor Vlek was sceptical, though, of the prospects. Producing enough food for projected populations of the developing world in 2020, for example, would mean supplying 185 Mt of added plant nutrients (nitrogen, phosphorous and porassium fertiliser) a year, compared with 62 Mt in 1990. To restore levels of plant nutrients in the soil, instead of maintaining it in its present depleted state, would mean adding not 185 Mt but 251Mt a year in 2020.

The explosive growth of cities has a major effect on plant nutrient budgets. A proportion (ideally, all!) of the nutrients end up in food, which is "exported" from the countryside to the city. Food ends up in sewage, most of which ends up in the sea. Recycling sewage as fertiliser is impossibly difficult to organise economically (it may happen in China, by decree). Worse, city sewage is very often contaminated with heavy metals from light industry. But recycling must in the long run be achieved.

Still more nutrients are exported to cities in firewood.

"Soil scientists tend to believe that soil is important," noted soil scientist Professor Gregory ¹³ "Are all soils important," he asked, "and are they important all of the time?"

Soil scientists have been rather successful in wiping out the differences, at least for farmers who can afford fertiliser and irrigation. But the old notion of a sack of fertiliser printed with a recommended dose won't do any more.

All the measures proposed for increasing and sustaining food production depend on fine-tuning agriculture almost hectare by hectare. There simply never will be enough soil scientists or meteorologists to make detailed recommendations on that scale. Everything depends on the farmers, and on the information they have to base decisions on. Professor Gregory speculates about "a sort of Sub-Saharan Archers" as a way of getting useful information out. The BBC radio soap opera "The Archers", set in the farming community of Ambridge, was launched on 1 January 1951 at the behest of the Ministry of Agriculture, as a means of informing and educating farmers in Britain.

A lot of that information will be fairly familiar to farmers, or their parents. Traditional small fields in England reflected the scale of variation of soil and water supply. Agroforestry is very similar to the forest gardens of Amazonia and New Guinea.

5. What shall we eat?

Whether our children can feed themselves depends on what they expect to feed themselves.

On the one hand, increased prosperity seems to be essential to the "demographic transition", and hence to achieving a stable population.

On the other, it massively increases demand for food production. Dr de Vries ⁷ calculates that a purely vegetarian diet requires 490kg of grain equivalent per person per year. A diet with a moderate amount of meat requires 860 kg/year; and an affluent, US-style diet 1500 kg/year. The reason is that it takes a lot of vegetable crops - primary food production - to feed animals for meat. Already, in South-East Asia increased affluence is leading to much higher demand for primary food production.

The growth of population alone means that primary food production must increase by a factor of about two. Changes in expectations - burgers for all! - could mean that there was demand for six times as much primary food in 2050 as is grown now.

The chances of such a sixfold increase in primary food production being achieved are, in scientists' language, "vanishingly small". In lay terms: "...and pigs might fly". Few of our children will have a high-meat diet.

6. Not by bread alone...

Dr de Vries also pointed out that there will be massive competition for land. As the remaining oil becomes more and more expensive to extract, there will be pressure to grow crops for fuel and not for food. You can run a car from 2 ha of well- managed land in many parts of the world - or you can feed a couple of families.

Fuel already competes with food. In parts of Ethiopia, Dr Sanchez¹⁰ said, almost every organic thing grown goes for fuel - not even for food and animal fodder. And Professor Vlek¹² observed that it's not easy to convince farmers in the tropics to leave straw in the fields as mulch to conserve water, if they have to build their houses from it and use it as fuel.

Economic development also competes with food production. One participant noted at lunch that in her country almost all the fertile land is in flat valley-bottoms; and this is where US companies are building their sprawling, low-rise micro- electronics factories. Another predicted that by 2050 the Philippines may have no Grade A land left.

7. Land conflict: how much may we use?

Dr Tinker¹⁶ quoted US computer millionaire Ross Perot as saying "If our children are hungry we will cut every last tree. And we will not worry about the Spotted Owls, except maybe to eat them."

Expanding food production competes for land with other species and cultures.

He asked "Why are people so excited about the destruction of the tropical forest?" He gave four reasons:

• Destruction of tropical forests implies a massive release of carbon dioxide into the atmosphere, increasing the risk of global warming;
• Felling the forests would severely change their local climates over a much shorter time-scale;
• The forests contain a very significant portion of the world's bidoversity; and
• Aesthetic considerations cannot be ignored.

(To that list this author would add the importance of the people who live in the forests and of allowing their unique human cultures to enter the 21st century on their own terms.)

If current trends continue, the entire rainforest of Zaire could have been felled by 2050. The tropical forest of South-East Asia may be largely lost by 2030. The position of "montane" (upland) forests is, however, worse than that of the rainforest; these are disappearing at 1.1% per year, compared to 0.6% per year for tropical forest in 1981-1990.

Dr Tinker is optimistic about farmers dealing with erosion on their own land. But it is in no-one's direct interest to deal with off-site effects - and so he is pessimistic about that. If you can put a price on these effects, something will be done about them.

Reafforestation and agroforestry, for example, have their off-site costs. They are designed to reduce run-off of rain and so to maximise the amount of water which passes through plants and into the air. That means less water for everyone downstream.

8. The shrinking world:
the frontier is closed

The liveliest session of an engrossing meeting was the half hour at the end devoted to general discussion. The major theme was one of increasing international interdependence.

It seems unlikely that China and the Punjab will in future achieve the same yield increases that they managed in the past 40 years. So will farmers in the Punjab get fed up with wheat and rice, because they're not making enough money? Will they diversify into high-value crops because the Indian middle class demands them - and is there then the possibility of Europe and the US becoming the grain-baskets of the world?

One participant predicted that "We in Africa will buy maize from Indiana, where they complain about 9 t/ha yields, and we'll sell you cures for prostate cancer and other high-value crops."

This notion depends, as Dr Bie¹⁷ pointed out, on transport infrastructure which does not exist - and which it is hard to see being built in the next 20-30 years.

9. What, then, is to be done?

More research is needed, to gain a better understanding of how food supply can be maximised with a minimum input of expensive resources and a minimum degradation of the soil and environment.

But the matter of anything being done about that research or its results is intensely political. The issue is not so much food supply, as poverty. "As scientists," one concluded, "we have a political and moral responsibility to make politicians aware of the issue."

Dr Bie, leaving the Food and Agriculture Organisation that month, noted that:

Governments sign many things. The Universal Declaration of Human Rights includes food as a human right and there are 840M people who are food-insecure. So some governments have not been doing their job. The World Food Summit made a commitment to halving the number by 2015; Fidel Castro said this was in a way disgraceful and that they should aim to feed everyone.

Grand-scale projects will not fulfil that aim.

It seems over-optimistic, for example, to assume that the the next 20-25 years will see enough economic development to provide farmers everywhere with fertilisers, credits to buy them and roads to deliver them. Such vast projects are simply not completed in that sort of time - unless, perhaps, they are awarded the overriding priority that the US and Soviet nuclear weapons projects had. To do so would mean confronting another concern which was fashionable in the 1960s, is urgent now and was outside the remit of the meeting: the finite nature of fossil fuel reserves. (A similar meeting on energy is indicated.)

Governments make less policy than they think they do, and the people who get soil on their hands make much more. Both scientific research and practical support for food production must be done in collaboration with these small farmers, and must help them make the best use of whatever resources they have - in many cases, little more than their own ingenuity.

Professor Gregory summed up: "By 2050 there will be eight billion people in the world. Those people will feed themselves - but only if we do the necessary research to increase crop production." Water is the worrying question mark over that prediction.

"They will," he continued, "do so mostly using existing technology. The population will stabilise - one way or another, unless it declines because they fail to feed themselves. And what will be the cost to the global environment?"

"Then we're left," Professor Gregory concluded, "with a very considerable management problem, to maintain that level of food production. In a sense, what we're talking about now is the shape of agriculture in 2075."

Notes and contacts

[1] Luke 8:26-33

[2] Professor D.J. Greenland F.R.S.

[3] Dr G.K. Heilig and Dr G. Fischer

[4] Dr E.B. Barbier

[5] Dr L.T. Evans, F.R.S.

[6] Dr M.V.K. Sivakumar

[7] Dr F.W.T. Penning de Vries

[8] Dr M.Falkenmark

[9] Dr J.S. Wallace

[10] Dr P.A. Sanchez

[11] Dr R. Webster

[12] Professor P.L.G. Vlek

[13] Professor P.J. Gregory

[14] Professor R. Lal

[15] Professor J.K. Syers

[16] Dr P.B.H. Tinker

[17] Dr S.W. Bie

[18] Throughout, "billion" is used in the common sense of one thousand million.

[19] One hectare (abbreviation: ha) is a square 100m on a side, one-hundredth of one square kilometre or, if you must, 2.47 acres. The prefix G stands for 1 billion; M stands for one million.

[20] Thomas Malthus, An Essay on the Principle of Population as it affects the future improvement of society with remarks on the speculations of Mr. Godwin, Mr. Condorcet and other writers: London, printed for J. Johnson, in St Paul's Church-Yard, 1798.

[21] Joel E. Cohen, How Many People can the Earth Support?: NY NY 1996 pub. Norton, ISBN 0-393 03862 9

[home]

Terms and license:

1. Population growth

2. Land

3. Who decides? Economics for small farmers

4. Land and crop productivity

5. What shall we eat?

6. Not by bread alone...

7. Land conflict: how much may we use?

8. The shrinking world

9. What, then, is to be done?