Saturday, November 26, 2016

Efficiency and productivity two very deceptive words

Dear readers,
I have been very busy with the launching of the Swedish version of Global Eating Disorder and some consultancies. Now I am soon going to Indonesia to look at palm oil production. Therefore I don't have time to produce much new material for the blog. I offer you a chapter of Global Eating Disorder below, discussing two of the most deceptive words I know. Efficiency and Productivity.

We often hear that ’productivity in farming has increased tremen­dously’, ‘the farmer of today is much more efficient than her ancestor’ and similar expressions. But those are deceptive words. They come with an air of being measurable and objective. And while they can be measured, the yardstick we choose to measure them against already predetermines the outcome. One has to also reflect over who benefits; what is productive or efficient differs depending on who asks the question.[1] Most choices are made on the basis of the profitability for the person or entity that has managed to, with whatever means, control that piece of nature.[2]
Throughout modern times the prevailing measure of grain produc­tivity in Europe was the ‘yield ratio’, i.e. the quantity that could be harvested from a certain quantity of seeds. This ratio was often in the range of five in Scandinavia and Germany, reaching ten under good conditions in England and the Low Countries.[3] For a grain-based culture, with access to rather abundant land and labor resources, this way of measuring makes some sense. And this also explains quite well why the Asian rice based cultures were more advanced than the Europeans, until the industrial revolution; the seed ratio of paddy rice is normally very high. Today, seed is only one of many inputs in farming and most farmers would find it strange to measure productiv­ity in terms of seed use even if it still prevails in some places. 
The size of the area unit itself was often dynamic and linked to the land’s ability to support people. Around the world, the amount of seed that had to be sown to feed a family from the harvest provided a basic measurement of the land. Even in the 20th century, fields in Guang­dong Province next to Hong Kong continued to be measured in dou, roughly equivalent to seven kg of rice seed, while up the mid-19th century land in New Mexico was computed by the fanega, approxi­mately fifty kg of wheat seed. I have also encountered land measurements based on how much land a man could ‘open’ in a days work, common in cultures with swiddening farming. And ’acre’ has similar roots; an acre was the amount of land that could be ploughed in one day with a pair of oxen. When land, labor and food become commodities, most of these localized and subjective measurements fell into disuse.[4]
These examples are good to remember when we discuss productiv­ity and efficiency. Productivity in farming can be measured in many ways. Per area unit; per person-hour; per unit of deployed capital; per energy input or; per water unit. The comparisons can consider total biological production, ecosystem services or only what is directly useful to human beings in the form of food, fiber and fuel. And the results, the useful resources, can also be expressed in different ways, for example, as kg (tons or bushels), as calories, as proteins or simply as money.[5] We can also ask if the productivity is serving to maintain the productive resources or if it is based on extraction of non-renew­able resources, resources which perhaps were abundant but now are increasingly scarce? Can we even talk about productivity if the produc­tion is based on the unsustainable use of irrigation, fossil fuel and soil management practices which erode the soil? 
We can also compare farming with ‘nature’. For example, the biologi­cal production (mass, energy or protein) in a farmed system can be compared with that in the natural system before farming. ‘Productiv­ity’ can have a completely different meaning if we counted the impact on ecosystems and the external costs caused by farming. Studies of a wetland in Canada, a forest in Cameroon and a Mangrove forests in Thailand showed the total value of these ecosystems was much higher than the value of the farming systems to which they were converted.[6] But, importantly, the benefits of the systems accrue to different people. 
In the long term, it is more interesting to increase the productivity relative to a resource that is limited rather than relative to a resource that is abundant. From this perspective, the contemporary obsession with labor productivity is strange, as we never had so many people on the planet and so limited natural resources. But if we look at monetary value we understand why we always try to save on labor as labor is costly and nature is ‘free’.

The total energy harvested per hectare can increase with increased use of ancillary energy; in many situations one can increase yield per hectare fivefold (or more) by using more energy. This energy can be used for better (and timelier) soil preparation, irrigation, fertilizers, etc. The ratio between energy output and energy input (i.e. efficiency in the use of energy) seems to be fairly constant up to a certain level, after which it rapidly deteriorates. Industrial farming systems, have long since passed the optimal energy use level.[7] According to FAO, industrial farming methods require 6,000 MJ of fossil energy (corre­sponding to a barrel of oil) to produce a ton of maize but traditional methods in Mexico only require 180 MJ (corresponding to 4.8 liters of oil) to produce the same. This calculation includes the energy needed for chemical fertilizers, irrigation and running machinery, but not the ‘shadow energy’, the energy used for making the machinery, for transporting products to and from the farm and for constructing the farm buildings. The energy ratio is below one – i.e. more energy is consumed than produced – for modern rice farming and just above one for modern maize farming, whereas traditional production of rice and maize gives a 60-70-fold return on energy used.[8] In light of this, to make statements that ‘productivity has increased tremendously in farming the last hundred years,’ is simply misleading. 
Yield per area unit has increased considerably, both in developed and developing countries. For example, in the United States, maize yield increased from around 3,700 kg per hectare in 1961 to around 9,500 kg in 2011. Wheat yields in Sweden tripled between 1900 and 2000. Global rice yields increased by one third in the twenty years between the end of the 1970s and the end of the 1990s. The most dramatic increases can be accomplished in highly regulated systems. In greenhouses with climate and water control, productivity per hectare is up to more than ten times that in an open field. The more advanced producers in the Netherlands take more than 60 kg of tomatoes per square meter, whereas open-field production may only reach some 5 kg per square meter. Still, the profitability of Dutch greenhouse tomato production is very low and largely rests on them getting very favorable gas prices. One should not forget that in many cases yield per hectare is low just because it doesn’t pay to increase it. 
Land price is a factor that greatly influences productivity per area unit. To some extent, it works both ways. Land with high yield poten­tial has a higher price than land with low potential and when you pay a lot for land you have to get a high yield per area unit. But in many countries and places, the correlation between land productivity and land prices is weak and land prices are influenced also by alternative uses (conversion to forest, industries, and housing), status and culture. 
Farming today is very capital intensive. In Sweden, it comes only after utilities and real estate in capital intensity; car-making has less than half the capital costs for a full-time job,[9] and in United States US$1.2 million is invested in each full-time agricultural job.[10] Yet at the same time the return on invested capital in farming is mostly low. The real return is often realized when the farm is sold. As the saying goes ’farmers live poor and die rich’: they reinvest most of their profits in their farms. For most farms, especially family farms, the concept of return on capital is somewhat alien. Families invest their capital and their labor on the same farm and there is no distinction in their way of thinking or in their books (if they have any) between the return on capital and the return on labor. One way to assess return on capital is to look at land rents. In the highly commercialized agricultural sector in the Nether­lands, land rents are administratively defined on a level of 2% of the calculated production value of the land. If this was not the case, if the farms had to render a commercial rent of say 4% of the land price, farming would not be profitable.[11] Land rents in the United States have fallen from around 7% of the productive value in the 1970s to a low of 3.4% in 2008.[12]
A lot of the support to farms in high-income countries is ’capital­ized’ in the form of higher land prices. This means that the subsidies increase the cost of the land and this makes it more difficult for new farmers to establish themselves. Depending on the nature of support other production factors can also be affected. In 2006, milk quotas (if you produce more you face a levy) in the Netherlands were valued at an astonishing €20 billion, whereas the gross annual value of milk production was around €3 billion. The right to produce milk thus became a commodity, more important than actual milk production[13] (due to changes in EU agriculture policies the prices of the quotas have dropped recently, and they are about to be abolished in the future). The right to direct support from the EU’s CAP (the ‘single farm payment entitlement’) has also become a tradable commodity. Buying and selling these rights has become a business in itself; ’buying an entitlement to a stream of future payments is an investment decision’, according to one analyst.[14]
Contrary to what many may believe, return on capital in farming in developing countries is high. There is very little available capital and even less is invested in farming, but that which is invested is often borrowed at rates of 20-40%, a return on capital which few investors in high-income countries can dream of. Despite this, international capital is not rushing to support the farming sectors in developing countries, at least not in the shape of credits to small farms. The risks involved are also high; farmers can be unreliable borrowers who often can get away with not paying their debts, or delaying payment for years.[15]

While a human being needs a few liters of water to drink, at least one thousand times as much water is used to produce food.[16] The water needed for different foods varies tremendously and even varies for the same product grown under different conditions. Often the figures used mix different kinds of water with no clear distinction. There is ‘blue’ water – water in rivers and lakes, ‘green’ water – water in rainfall and in the soil, and ‘grey’ water – the water that is needed to absorb or purify the waste. Together these make up our ’water footprint’. The Water Footprint Network broke down the water footprint of a mar­gherita pizza (one topped with tomato, mozzarella and basil). They found that it takes 333 gallons (1,260 liters) of water, enough to fill almost ten bathtubs to make a single pizza.[17]
But statistics can be presented in many ways. The water footprint of beef is big, ten times bigger than for grain and fifty times bigger than for (some) vegetables. On the other hand it is smaller than the water footprint of sesame oil, olive oil, coffee, cocoa and a number of other crops. Rainfall is also not the same as the water in lakes or wells. In all dryland areas, except where there are irrigation possibilities, livestock keeping has been the traditional way to get food from the land. And this for very obvious reasons; it is more water-efficient to raise live­stock than to try to grow crops that don’t get enough water. In Namibia – one of the driest countries on the planet – most of the water falls on the dry rangeland as rain and only a small part is supplied as drinking water for the animals. A borehole for cattle with a solar pump and a capacity of 2 m3 per hour (20 m3 per day) can service some 250 cattle. They will produce in the range of 20 tons of meat per year. The same quantity of water would be enough to irrigate approximately a hectare of farm land, which would produce considerable less food (say 3-7 tons of maize) and a lot less money. If we calculate the total water use of the land and include the rainfall, we arrive at a high water foot print. But this rangeland cannot be used for cropping as the rainfall is to low. The only other alternative use of the land would be to have game roaming there.[18] Therefore, we cannot conclude that cattle-breeding, under those conditions is wasteful of water. 
Also, water footprint per kg is a dubious measurement. We eat very different quantities of food, and the concentration of useful nutrients that they contain differs widely. As an example, we have to eat 10 kg of broccoli for our daily calorific needs, but only 0.5 kg of bacon. If we look at the use of blue and green water per calorie, nuts have the lowest efficiency in water use and root crops the highest, according to research­ers Mekonnen and Hoekstra. Most vegetables and meat rank quite similarly. In terms of blue and grey water use for protein, oil crops are the best, followed by milk, lamb, root crops and grains. Nuts and vegetables are the least efficient.[19] Seen on a larger scale, attempts to increase yield per hectare mostly also increase water productivity. An unirrigated hectare of wheat uses more or less the same quantity of water when it produces 1500 kg as when it produces 5000 kg.[20] If irrigation is introduced into the sys­tems, the effect is, perhaps sur­prisingly, that water productivity increases; that is, the increased yield resulting from irrigation uses less water per kilogram than the original yield based on rainwater. In a situation where the annual rainfall is around 400 mm, water stress poses a real limitation to crops, adding some 100 mm of water at critical moments can easily double yields.[21]. According to researchers at the Stockholm Environment Institute there is a large potential to improve water productivity through using improved, existing, water management practices. They predict that, globally, we will need to increase water usage by around two thirds of present levels (7000 km3/yr) in order to feed the world in 2050 but that water produc­tivity improvements could save up half of that quantity.[22]

Because labor costs represent a very high share of farm costs, pro­ductiv­ity per person-hour or per person-day has been central in leading to the increased mechanization of farming. Labor productivity is also very important from a broader societal perspective as increased productivity of agricultural has released people to work in other sectors. Industrialization could not have taken place unless labor had been released from farming. One person in the United States occupied in farming can now produce food for some 300 people, in some devel­oping countries one farm worker can only feed a few. Another way of looking at labor productivity is to see how much grain can be pro­duced per person-year. Grains are the most important foods in the world, are grown in most places and as such are used as a proxy for general agriculture development. In the areas with lowest productiv­ity, one person can produce not even 1 ton of grain per year, whereas the most productive farms produce up to 5000 tons per person-year as is the case on Bob Stewart’s farm in Illinois. 
Of course, these comparisons mask the fact that a modern farmer doesn’t even ‘feed’ him or herself or the family or their animals. A modern farm is a company where other people’s goods and services are bought and sold. There are accountants, repairmen, consultants, machine operators, computer service technicians and meteorologists, all in the service of the modern farmer. All this work is embedded in the farm produce. Earlier generations of farmers, and still many farmers in developing countries, produced not only food and fodder for themselves and their animals, but also fiber for clothing, leather for shoes and most of their own tools and medicines from their farms. But even when we have taken that into account, the difference in labor productivity between farms in high-income and low-income countries is still staggering. Energy use is the single most important determining factor here as there is a very strong correlation between the energy resources commanded by one person and that person’s productivity.
Productivity, measured in added value per worker, is quite natu­rally much higher in high-income countries than in developing countries. For instance, a farmer or farm worker in Malawi produced an added value of just US$130 per annum in 2001-2003, a Chinese farmer US$368 while their French and American colleagues produced US$39,000 and US$36,000 respectively. Even more worrying is that this gap is widening. The OECD countries had an agricultural labor productiv­ity that was 25 times higher than that in Sub-Saharan Africa in 1971. By 2005 this had increased to 68 times.[23] A report to the FAO projects that revenue per agriculture worker in Sub-Saharan Africa will increase by a meager 50% between 2010 and 2050. Growth in productivity in South Asia (India and its neighbors) will also be slow, and Latin America is the only region likely to catch up with developments in the OECD countries.[24]
Those figures come close to what farmers earn and give an idea of what farm workers are – and can be – paid. It works both ways; if workers are productive they can get a good salary which in turn means that there are incentives for saving labor, which in turn means that income per worker will increase. In addition, if wages in other sectors are high, agriculture wages will follow (even if they mostly stay comparatively low). A particularity of farm labor is that, in most parts of the world, it is markedly seasonal. When it is too cold, to dry, too hot or too wet, there is simply not much to be done in the fields. This means that farm labor often is underemployed, or that seasonal workers take up the jobs. Farm work is also considered to be ‘un­skilled’ even though I personally object to this categorization. In my former role as farmer and employer I can assure the reader that the productivity of people weeding or making bundles of parsley can easily differ by a factor of three or four, depending on their skills. Taking good care of animals also requires a huge amount of skill, know-how and ability. And modern farm work is certainly highly sophisticated. 
It is somewhat puzzling why most agronomists and institutions focus so much on yields of crops per hectare as the main measure of agriculture productivity, when in reality that is not a driving force for farmers who look more at productivity per labor unit, or if they are modern agri-business operations, the productivity of capital invested. If we compare farms globally the farms with the highest yields per hectare are rarely the most competitive. European farmers generally have much higher yields per hectare of wheat than their Argentinean, American or Australian colleagues, still they cannot compete and are dependent on support programs of the European Union, because their general cost levels are higher. Similarly in the dairy sector, the world market is dominated by a country with a low milk yield per cow. The dairy industry in New Zealand is still primarily built on grazing cows and production per cow is low by international standards. The average production per cow in Israel was 12,500 kg in 2007, while it was less than 4,000 kg in New Zealand,[25] but New Zealanders produce milk a lot cheaper than Israelis. 
If we compare efficiency in various systems, e.g. in farming or food processing, in most cases this will show that the bigger and more technological advanced system is more competitive. But are they more efficient and productive? Often, small farms have a higher yield per hectare than large farms, but large farms are still gradually squeezing smaller farms out of the market, because of market access, possibilities for rational specialization, economies of scale, better access to credits or governmental policy distor­tions.[26] Larger crop farms perform better financially, on average, than smaller farms. This is not because the larger farms have higher revenues or yields per area unit, but because they have lower costs. As expressed in the report Farm Size and the Organization of U.S. Crop Farming from USDA: “larger farms appear to be able to realize more production per unit of labor and capital. These financial advantages have persisted over time, which suggests that shifts of production to larger crop farms will likely continue in the future.” Their yield per hectare is mostly the same as on smaller farms but USDA’s research shows that farms with more than 2,000 acres spend 2.7 hours of work per acre of maize and have equipment costs of US$432, while a farmer less than 249 acres will spend more than four times as much labor and twice the amount for equipment per acre. In this limited sense the larger farms are indeed more ’efficient’ and ’produc­tive’.

There are many different ways to look at farm productivity and, depending on what and how we measure, we can draw different conclusions. In principle, the factor which is scarcest will, and should, be the most important. Farms in high-income countries are character­ized by a high input of energy and a low input of human labor. They have no shortage of labor but it is costly and therefore productivity per work-hour has been the strongest driver of change. Close to cities, or in very densely populated areas, land is scarce and farms are more shaped by high land prices. At a certain land price, grain farming is no longer viable and farming will move towards higher value crops, or will become a playground for the rich, golf courses or paddocks for race horses. 
Economists talk about ‘total factor productivity’, a rather opaque measure which has a scientific air. It does sound like a good idea to combine all the factors of production in one measure. But as this is measured in monetary terms it will just value things by their market value. So if labor is 200 times more expensive in one country than in another you have to produce 200 times more per hour to achieve the same productivity. And if water is free, water productivity will not be reflected at all. In this way, productivity comes to mean more or less the same as profitability and becomes a circular form of reasoning that is of little value in discussing the big picture, even if it does reflect quite well what guides a modern commercial farmer. 
We need to redefine productivity. But it is not sufficient to theoreti­cally redefine productivity, we also need to redesign the economic system which has created a distorted view of what is productive and what is not. Today, productivity is measured by how many trees one person can cut down with her chainsaw or how much fish a fisherman can scoop up from the sea. But as natural resources dwindle, the real productivity lies in how these resources re-generate. We are produc­tive if there is more forest next year than today, if there are more fish and if the soil becomes more fertile by the years instead of being exhausted and eroded. In a similar way we are efficient if the food we produce and consume is healthy rather than if it is cheap. 

You can read more blog posts on this theme, e.g.
Jevons paradox - why efficiency is a liar wordHow increased labour efficiency drives resource consumption

[1]           Rundgren, G. 2013 Garden Earth - from hunter and gatherers to global capitalism and thereafter Regeneration.
[2]           Millenium Ecosystem Assessment 2005 Millennium Ecosystem Assessment: Ecosystems and human well-being–synthesis World Resource Institute.
[3]           Grigg, D. 1983 The Dynamics of Agricultural Change Palgrave Macmillan.
[4]           Linklater, A 2013 ‘When Pilgrims Privatized America’

[5]           Rundgren, G. 2013 Garden Earth - from hunter and gatherers to global capitalism and thereafter Regeneration.
[6]           Millenium Ecosystem Assessment 2005 Millennium Ecosystem Assessment: Ecosystems and human well-being–synthesis World Resource Institute.
[7]           Bayliss-Smith, T. P. 1982 The Ecology of Agricultural Systems Cambridge University Press.
[8]           FAO 2000 ‘The Energy and Agriculture Nexus’ Environment and Natural Resources Working Paper No. 4. United Nations Food and Agriculture Organization.
[9]           Malmaeus, M. 2013 Tillväxt Till Varje Pris? Notis.
[10]          Schnepf, R. 2014 US Farm Income United States Congressional Research Service.
[11]          Ploeg, J. D. van der 2009 The New Peasantries: Struggles for autonomy and sustainability in the era of empire and globalization Earthscan.
[12]          Farmdoc 2010 ‘Farmland Price Outlook: Are farmland prices too high relative to returns and interest rates?’ 18 October 2010, University of Illinois.
[13]          Ploeg, J. D. van der 2009 The New Peasantries: struggles for autonomy and sustainability in the era of empire and globalization Earthscan.
[14] 2013 ‘CAP Reform Uncertainty and the Market for Entitlements’
[15]          Rundgren, G. 2013 Garden Earth - from hunter and gatherers to global capitalism and thereafter Regeneration.
[16]          Kijne, J. et al. 2009 Opportunities to Increase Water Productivity in Agriculture with Special Reference to Africa and South Asia Stockholm Environment Institute.
[17]          Grace Communications 2013 Food, Water and Energy, Know the Nexus. Grace Communications.
[18]          Meyer von Bremen, A-H. and G. Rundgren 2012 Jorden vi äter, Swedish Society for Nature Conservation.
[19]          Mekonnen, M.M. and Hoekstra A.Y. 2012 ‘A global assessment of the water, footprint of farm animal products’ Ecosystems (2012) 15: 401–415.
[20]          Rundgren, G. 2013 Garden Earth - from hunter and gatherers to global capitalism and thereafter Regeneration.
[21]          Renault, D. 2002 Value of Virtual Water in Food: Principles and virtues United Nations Food and Agriculture Organization.
[22]          Kijne, J. et al. 2009 Opportunities to Increase Water Productivity in Agriculture with Special Reference to Africa and South Asia Stockholm Environment Institute.
[23]          Sauber, M. 2010 ‘Low agricultural productivity in a monetary economy’ paper to the 12th Annual conference of the AHE Association for Heterodox Economics.
[24]          Schmidhuber J., J. Bruinsma and G. Boedeker 2009 ’Capital requirements for agriculture in developing countries to 2050’, Paper to Expert Meeting on How to Feed the World in 2050 United Nations Food and Agriculture Organization.
[25]          Gerosa, S. and J. Skoet 2012 ‘Milk availability, trends in production and demand and medium-term outlook’ FAO, ESA Working paper No. 12-01.
[26]          Rundgren, G. 2013 Garden Earth - from hunter and gatherers to global capitalism and thereafter Regeneration.