Eten van eigen bodem : een modelstudie

Introduction and object of the study

Since the great agricultural crisis at the end of the 19th century, the traditional bond between animal husbandry and crop husbandry in Dutch agriculture has been gradually broken. The enduring availability of cheap cereals (and cereal-substitutes) since then, enabled the rise of a huge and highly specialised dairy- and meatsector, which nowadays is one of the predominant characteristics of agriculture in the Netherlands.
This specialisation-process has not been entirely without interruptions. During the last years of World War I, the years of the Great Depression and the following Occupancy by Germany, there were - inevitably - lapses. After 1945, however, the process of Dutch specialisation in animal husbandry-products regained its momentum, not in the least owing to the better perspectives of a growing Common Market after the Dutch joining of the European Community. Thanks to this Common Market, Dutch agriculture could play an important role in the welfare of the national economy - directly and indirectly: by its contribution to balance-of-payments surplusses. In terms of value, the Netherlands evolved to one of the world's leading net-exporters of food products.
This rather comforting position, however, leans heavily on the Import of foreign feed-stuffs. Besides, it is also highly dependent on the availability of fossil energy: for the production of fertilizers, running the machinery, etc. To give an impression of these dependencies, the following figures are rather illustrative: In 1980 Dutch animal husbandry not only laid claim to the products of 75% of our own agricultural soils (2,000,000 ha), it also consumed the products of an equivalent of about 5,400,000 ha cereal-land in the USA. At the same time, our own land resources are exploited more and more intensily, by using large quantities of chemical fertilizers. For instance: nitrogen fertilization In 1950 amounted to just 156,000 tons (N-content), but that figure rose to 220,000 tons in 1960, 400,000 tons in 1970 and 490,000 tons in 1980 - notwithstanding the simultaneously increasing supply of animal manure, from the rapidly expanding herds of dairy cows, fattening calves, pigs, laying hens and broilers. In this same period, population In the Netherlands swelled from a mere 10 million people in 1950 up to more than 14 million souls in 1980: i.e. an increase of over 40%.
Seen from this angle, there certainly seems reason to doubt the capacity of the Dutch agricultural sector to keep up with national food-demands if - in the event of a grave and lasting international crisis - it no longer can rely on its huge import
lifelines. It may seem a little bit farfetched to inquire into such a remote possibility since our national and international interests seem properly secured by the Dutch bonds with NATO and EC. Be it then only for the purposes of a closer insight in Dutch civil defence possibilities, it nevertheless is interesting to know what chances there are for the food-supply in such an autarchic situation. This, then, was the object of the study.

Retrospect and prospect

Remote as it nowadays may seem, since the beginning of the 20th century the Netherlands had actually twice to face such an autarchic situation. Firstly in the last two years of our neutrality during the Great War, and later on during the entanglement of the German occupational forces in Holland in World War II. The Dutch experiences in both periods were rather bitter. Especially the hunger-winter of 1944/45 has made a deep impression - and still does, in the minds of its survivors.
To be fair: that winter of 1944/45 was not strictly the result of a national isolation. It was a complex of circumstances that led to famine in the cities in the West of Holland: a rather static military front-line blocked food transports from the South, transport in still occupated areas was hampered by a severe and lasting strike of railway-workers, occupational forces claimed priority for their food-supply, etc.
Such specific "internal" difficulties did not occur in the autarchic period of 1917-1918. Nevertheless - had there be no armistice in the autumn of 1918, a hunger-winter 1918/1919 would no doubt have occurred as a result of the lasting blockade of Dutch sea-routes: we escaped really on the verge of it.
Thanks to the technological and biological development agriculture has experienced during the last decades, it seems possible to cope with such threats in future. Even without the Import of raw feed (and food) materials, we at present reap from our own soils enough edible calories to feed every member of the Dutch society: 2,400 calories per person per day. That is: if we are able to keep up with the energy-demands of the productionprocesses and if consumers do not mind the sweetness of this calorie-package for the time being, as almost 30% of the edible calories are sugar-calories!

Modelling the autarchic food supply

This, of course, cannot really be a lasting perspective for an autarchic food- supply. First of all, the caloric availability should be more in line with nutritional standards. And secondly, there is reason to believe that an autarchic situation will also lead to a severe Internal scarcity of energy-intensive means of production: fuel-products, artificial fertilizers, etc. Anyway, energy-scarcity was a major problem in both 1917/18 and 1944145.
How, then, can the national food-supply-system be so reorganized as to meet both the demand for a more balanced national diet as well as a (supposedly) emerging energy- scarcity, when there Is only our own 2,000,000 ha land- resources to fall back upon? For answering this question, a linear programming model has been designed, that encompasses all land- and energy-using components of the autarchic food system: production of artificial fertilizers, tillage-activities, animal conversion, industrial processing of agricultural products, transport-activities and consumption of food products.
The model gives a detailed technical representation of the relationships between the activities that constitute these components of the system. By means of the linear programming technique, those activities can be steered in such a way that the system Is optimized in regard to a specific objective as, for instance, the minimazation of the total energy-use generated by the food production, given dietary requirements. Furthermore, as those activities in the model are laid out not only to their physical form (growing of wheat, fattening of pigs, etc.) but also to the intensity in which they are performed (use of fertilizers, composition of feed-rations, etc.) and the location on which they are projected (agricultural areas, consumption-regions), the output of such an optimization procedure throws light on WHAT, HOW and WHERE the autarchic food system will be different from the system we know now.
We have run this model for four different scenario's, which will be described hereafter.

Self-sufficiency and minimal nutritional demands

First of all, we tried to make a reconnaissance of the foodsupply-system when it is organized to be as economical as possible with Its energy-use and only faces minimal dietary requirements. Those requirements, by the way, were formulated by nutritional experts of the Dutch Ministry of Agriculture and were roughly based on the same recommendations as the well- known US Dietary Goals: per person per day a quantity of 2,350 edible Kcal has to be available, minimum 12% of them being protein calories, maximum 10% sugar-calories and maximum 30% fat-calories. Furthermore, the fat-package must contain enough poly-unsaturated fatacids: 10 grammes per person per day.
Given these requirements, the model unveiled that there is ample room to survive, for only 52% of the agricultural area will have to be used (it is supposed that production-techniques will be the same as they are now as falling back on the manual labour of man and horse is Impossible). Cultivating more area or lower yields would induce more energy-use. The corresponding agricultural production is rather monotonous: only cereals, oilseeds and leguminous plants (clovers) are cropped - the latter only for the purpose of economizing the use of nitrogen fertilizers. Animal husbandry does not occur, with the exception of a minimal herd of fattening pigs, which is fed mainly with by-products of cerealand oilseeds-processing industries.
Consumers are accordingly faced with a very sober diet: a kind of water-and-bread-ration, with a little vegetable fat and a minimal slice of meat - some "enrichment" could still come from products which are not envisaged by the model: fish, vegetables, because of their marginal caloric contribution.
The sunny side of such a national diet is its low "cost" in terms of energy-use: only 1 milliard natural-gas-equivalents (31,8 MJ per n.g.-equivalent). The energy input- output-ratio which compares the Input of fossil calories with the output of edible ones, is highly positive: 61-100 for the whole foodsupply-system and 26-100 for the primary production, i.e. agriculture and fabrication of fertilizers. At present, the primary production in the Netherlands shows a negative energy-input-output-ratio of 125-100; this is exclusive of that part of the production which is based on imported raw materials.
How sunny it may seem in regard to Its "cost", such a monotonous agricultural production and food consumption are of course more an academic than a real perspective for an autarchic foodsupply-system. Nevertheless, this model-scenario had it purposes, be it only for its quantitative advertisement of the possibilities to alternatively arrange our food-system.

Self-sufficiency at maximum capacity

The next step, then, was to Investigate how far nutritional demands can be expanded before the limited availability of arable land in the Netherlands becomes an absolute bottleneck. And, furthermore, how such an expansion of food production would affect the energy-demand of the food-system.
This new objective results, of course, in a 100% use of all agricultural areas, on which new crops as potatoes, sugarbeets and grass (for pasture) are grown next to (not Instead of!) the cereal- and oil-seed-crops which were already present in the former scenario. All crops are fertilized as heavily as possible, to ensure the highest possible physical returns.
Animal husbandry, too, now comes clearly Into the picture, not only as pig-husbandry but - predominantly - in the form of a considerable stock of dairy-cows: 1 million of these cows are presenting the Dutch consumers with roughly 5 million tons of
milk per year. Their feed-ration, by the way, consists not only of grass (85%), but also for a considerable part of home-made replacements of feed-concentrates: beets, turnips and by- products of processing industries.
If the corresponding food-package would be solely put at the disposal of the Dutch inhabitants, each of them could consume a varied calory-package of 5.100 edible Kcal per day. Now, of course, no real person - with a few exceptions - can cope with such an abundance of food-calories without harming his health or becoming extremely wasteful with foodstuffs. But we can place this result also in another context: if need be, the Netherlands can feed about 30 million people - twice the quantity of inhabitants It has now!
Meanwhile, there Is a "cost" to such an extremely high foodproduction: it usurps an amount of fossil energy that is more than five-fold the amount that was calculated for minimal nutritional demands: 5,5 milliard natural-gas-equivalents.
Both the abundance and the high energy-"cost" make such a self-sufficiency-at-maximum-capacity highly unlikely: this concept is also a purely academic one, that only serves to throw light on the other end of the food-supply-elasticity.

Keeping up with nutritionists: a proposed national diet

For nutritionists In the Netherlands, the present abundance with food-products has also a number of drawbacks. They oppose
the high caloric content of the food package in general, and the overrepresentation of fat- and sugar-calories in particular. For
our model-calculations, we asked some nutritionists to present us with a national food-package that would be more in line with
their views. This resulted in their following proposal:
- a standard food-package contains no more than 2,350 edible Kcal per person per day, of which:
- 43% stem from consumption of cereal- products;
- 17% from milk-products (excl. butter) and cheese (3%);
- 16% from edible fat-products: butter and margarine;
- 15% from potatoes;
- 4% from sugar-products;
- 4% from meat-products;
- 1% from eggs; and
- 1% from peas and beans.
This food-package, then, had to be produced at a minimal cost of fossil energy-inputs.
The model-calculations showed, that even now not all the available agricultural area is exploited: still 24% of our landresources are kept idle, i.e. lie fallow. As in the former scenario's, cereals and oil-seeds are fairly well represented: together they lay claim to 33% of the total land-availability.
Potatoes and sugarbeets - today's main crops in the Netherlands are far less important in respect to their claim on arable land. 8% and 1% respectively. The rest of the land-resources - i.e. 36% - is reserved for the cropping of feed-products, mainly grass and fodder-beets. All crops are dunged mainly with animal manure. Furthermore, overall fertilization-levels are rather low, extremely low even In respect to nitrogen- fertilization.
Although, In the end, more than half of the agricultural area is exploited to sustain animal husbandry-activities: 0.9 million dairy cows, 0.5 million pigs (yearly delivered) and 5.6 million laying hens - the contribution of these activities to the overall production of edible calories is fairly modest: 24%. Furthermore, by stressing the need for luxurious products as milk, cheese, meat and eggs in the food-package, the energy-cost of its production Increases from a mere 1 milliard natural gasequivalents In the former minimal-demands-scenario to over 1.7 milliard n.g.-equivalents now. Alternatively, this increase in energy-input could also have been used to produce more, instead of more luxurious, food products: It has been estimated that with an extra 0.7 milliard n.g.-equivalents and all Dutch inhabitants fed at a minimal nutritioal-requirements level, an extra 2.8-4.1 million foreign people can be supplied with a similar food-package.

Keeping up with our normal preferences: a bit over-eating

To make the picture complete, we finally confronted the model with the food-package we are consuming today: too fat, too sweet, too much stress on animal calories, but, all the same, quite In accordance to what we are actually used to. Can we produce such a package even in autarchic conditions? And, if so, what inputs of fossil energy-products would it, at the least, require?
The answer that evolved from the model-calculations was: yes, we are able to produce such a package, even in autarchic conditions, and even without exploiting all available land- resources. Compared to the former nutritionists' scenario, roughly the same pattern emerges in respect to the specific crops that have to be grown: cereals and oil-seeds are the predominant tillage-activities, accompanied - that is new - by pulses (peas and beans). Less important - in respect, that is, to the areas they claim - are sugarbeets and potatoes. Grass, fodder-beets and silo-maize are the main feed crops, and lay claim to about half of the available land-resources.
Animal husbandry-activities: 1 million dairy cows, 0.6 million fattening cattle (yearly delivered), 4.8 million pigs (idem), 80 million broilers (dito) and 9 million laying-hens, in fact claim more than that, as the main part of the cereals and pulses that are grown, disappear Into animal bellies.
The costs of all this In terms of energy-inputs are rather high: 2.3 milliard natural gas equivalents yearly. With the same amount of energy-input we alternatively could, if Dutch consumers would be willing to put up with a minimal diet, produce food for another extra 4.9-6.7 million foreign people.

Conclusions

If, in future, the Netherlands would be confronted with autarchic conditions, there is no need to fear mass-starvation. Our own soils, thanks to the intensive fertilization of the crops they grow now, produce enough edible calories to survive. Never theless, if such autarchic conditions last, It will be wise to reorganize our food-supply-system. Our model-calculations showed that there are plenty of possibilities for such a reorganization, as the availability of land is, in fact, no major bottleneck. If need be, the Netherlands can produce 120% more food products than is necessary for their own subsistence, though that would doubtless induce a considerable (opportunity) cost in terms of fossil energy-input. It is shown that it is really the national willingness/preparedness to pay for those costs, that determines the actual form of an adapted food-supply-system. Whatever the outcome, in advance it can be stated that such a reorganized system is different from the system we know now, in the following respects:
- cereals, oil-seeds and leguminous crops will push away potatoes and sugarbeets as major crops;
- food-crops are prevalent over feed-crops. This means that the pasture-area of 1.3 million hectares we know now, has to be diminished to the mere 0.6 million hectares which, be cause of their characteristic soils, cannot be used for tillage-activities;
- as a consequence, animal husbandry-activities will shrivel to the levels that are just necessary to supply the Dutch consumers with the animal calories they require. Those requirements - by the way - may well come under severe pressure of the authorities to economize the use of our national fossil energy reserves.
- in any event, dairy cow herds have the best credentials to be spared annihilation by feed-shortages. As cows graze pasturelands that cannot be exploited for food-crops and as they can be partially fed on non-edible by-products of the arable production, there Is still room for about 1 million of them;
- herds of pigs, broilers and laying hens, on the other hand, had better be drastically diminished as soon as possible, as animal calories and proteins in dairy products are far cheaper produced.
- as to the intensity of agricultural production: fertilization-levels won't really have to be as high as at present and emphasis should be laid on the recycling of natural sources as animal deposits and leguminous plants (clovers). In animal husbandry, the use of concentrated feed products should be strongly diminished, though not quite abolished as these concentrates are - for dairy cows - necessary to maintain a reasonable milk-production per cow, or - for other husbandry-activities - indispensable for the production of (pig/chicken)meat and eggs at the minimum- levels required. Such concentrates are preferably not edible cereal-products, but by-products from milling industries, fodder-beets, beetpulp and - if need be - pulses.

By running our model we were able to draft a sketch of the possibilities that avail for an autarchic food-supply-system In the Netherlands. When these possibilities are thus outlined, other questions may emerge as, for instance: how to realize one of those possibilities and what role is there for a dynamic government-attendance to the structural changes in production, processing and consumption that have to be pursued? Here we actually leave the boundaries of our own problem and enter the realms of the next one - which, in my opinion, would certainly be worth while a separate inquiry.

Supplement: mathematical framework, database and methodological background of the model

Thus far, attention was given to the results of the modelcalculations, more than to the model itself. Therefore, to give a complete picture of the research-activities we carried out, the mathematical structure of the model and the specific elements of its data-base are presented in a separate supplement. Furthermore, the methodological background of the linear programming technique, i.e. the Simplex Method, is more closely treated in this supplement, to perform a sensitivity- analysis on the results of the scenario-calculations. We analysed these results for two main features: how sensitive are they for the specific values we attached to the coëfficients of objective function and how are these results influenced by changes in the availability of major inputs to the foodsupply- system or by changes in the specific nutritional requirements that the system has to fulfill? The conclusion of this analysis was that, though the details of the model-results may be rather sensitive to such changes, their broader features are in fact quite robust.