Risks to Brazilian agriculture

19 Nov 2012 by Jim Fickett; minor correction 14 Dec 2012 (7% irrigated should have been 4%)

In the short run, at least, neither land nor water seems to be a serious limitation on the growth of Brazilian agriculture. The biggest risk to growth is infrastructure, with transport costing more than it should, hurting competitiveness. Other significant risks include changes in government policy and climate change.

As described in a previous post, Brazil's exports of unprocessed food have grown by more than a factor of four in the last decade or so (Growth in global demand for Brazilian farm products has been robust to slowdowns). Exports have grown more than farm production overall, but production growth has also been strong. For example, total production of grains grew from 58 million metric tons in 1990-91 to 144 million in 2008-09.

Every source one reads has a different estimate of the amount of land still available to put under cultivation. But all agree that substantial land is still available. In 2010 the Economist wrote:

Brazil has more spare farmland than any other country. The FAO puts its total potential arable land at over 400m hectares; only 50m is being used. Brazilian official figures put the available land somewhat lower, at 300m hectares. Either way, it is a vast amount. On the FAO’s figures, Brazil has as much spare farmland as the next two countries together (Russia and America). It is often accused of levelling the rainforest to create its farms, but hardly any of this new land lies in Amazonia; most is cerrado.

As far as preservation of existing farmland goes, a group of American farmers touring Brazil were impressed with the uniform high value Brazilians place on conservation. In particular, no-till farming is widely used to reduce erosion (see Bonds, commodities, Grantham, and the future of investing for more background on no-till). Thus it appears that land is not a near-term limit to further growth in Brazilian agriculture.

Brazil is also fortunate to have a great deal of land with excellent rainfall. Only about 4% of Brazil's land under cultivation is equipped for irrigation. Thus, unlike many countries where increasing water shortages look to be a serious problem (the US and China come to mind), it does not seem that water is a major limitation for farm output in the near future.

In reading about the farm sector in Brazil, probably the most frequent negative one encounters is infrastructure – transport from farm to city, or farm to port, is slow and expensive. Here is an excerpt from one report on a recent agribusiness conference:

Brazil must rid itself of this notion that it has a captive market at its feet, leading figures told an agribusiness conference in Sao Paulo.

“I have no doubt that China will continue to consume more soybeans and chicken; my doubt is who will produce these soybeans and this chicken,” Anderson Galvao Gomes, CEO at the Celeres farm consultancy declared.

At the heart of the issue is the country's inability to resolve infrastructure problems and reduce the regulatory risks, particularly in relation to foreign investments.

These obstacles are prompting investors to look seriously at alternatives in Africa, Eastern Europe and further afield, leaders said.

… the use of advanced technology means Brazilian soybean farming is still competitive with the U.S. inside the farm gate.

It's outside the farm gate that the problems really start. With new farms at least 300 miles from port, you need efficient logistics.

Unfortunately, they are lousy, with 60% of soybeans still transported by road – the least efficient way of going about it.

New logistics are coming on line, specifically rail links, but a lack of competition on these routes means they will offer only slim cost savings for farmers.

“In Brazil, we only see a 5% to 10% cost saving on rail over road transport, whereas in the U.S. you are talking about a 30% saving,” said SLC's Moura.

For competitive exports, the effect of poor transport is highly significant. It is hard to find quantitative data that is sector wide, but there are many examples that illustrate the problem. Here is one:

The cost of Mendonça's haul [Mato Grosso to Santos] amounted to nearly 40 percent of what the 37 metric tonnes (40.7 tons) of corn sold for in Santos. Transport across a similar distance in the United States, mostly by barge, amounts to only 10 percent of the price of U.S. corn at port.

Goods can also take three times as long to move a given distance as they do in China, a country that has used its run of economic success to invest heavily in roads, rail and ports. …

the low-cost advantage that Brazil once enjoyed is succumbing to rising transport costs. The jaunt from farm to port in Brazil already costs more than twice the sea freight fees to China, and that ratio is about to climb sharply as wages rise and the laws on rest periods for drivers take effect.

Also, although transport of farm product to end users is probably the biggest issue, it is not the only one. most fertilizer has to be imported, and delays at ports raise costs – or even introduce the risk of having to do without at times.

Government policy has been a major influence on Brazilian farm output.

  • Total output languished for decades due to policies that discouraged investment.
  • The green revolution in farm productivity was mainly due to a government research institution and extension service, Embrapa.
  • Total output grew quickly after hyperinflation ended in the 1990s.
  • Exports increased greatly beginning with a major policy shift that retracted many measures inhibiting trade.

When looking over the major shifts in policy that changed the outlook greatly several times in the past, it is clear that future changes could either help or hurt output immensely. In particular, the government budget is completely unsustainable, and high inflation is a significant danger. High inflation would specifically hit farming because it would negatively affect exchange rates, and hence the cost of fertilizer.

Climate change might be a major risk. I've never been able to find a realistic appraisal of how good the models are that predict the local effects of future global warming. But climate change is real and, for what it is worth, one major model predicts that Brazil could become much drier, with the Amazon forest converting to savannah. Even if this is in Brazil's future, it is difficult to know on what timescale drier conditions might have practical effects, though one study predicts significant effects in the arid northeast beginning in 2025, and another foresees significant losses already in 2020. For me the takeaway is that any investor in farming, in Brazil or anywhere else, had better become familiar with climate models and try to make a pragmatic assessment of the real risk.