Ed Dolan's Econ Blog

Does Peak Phosphate Spell Doom for Humanity, or Will the Market Save Us?

Although climate change catches the headlines, it is not the only doomsday scenario out there. A smaller but no less fervent band of worriers think that peak phosphate—a catastrophic decline in output of an essential fertilizer—will get us first.

One of the worriers is Jeremy Grantham of the global investment management firm GMO. Grantham foresees a coming crash of the earth’s population from a projected 10 billion to no more than 1.5 billion. He thinks the rest of humanity will starve to death because we are running out of phosphate fertilizer.  This post on Business Insider from late last year provides an array of alarming charts to back up his warning.

Foreign Policy agrees that phosphate shortages are a potential threat. “If we fail to meet this challenge,” write contributors James Elser and Stuart White, “humanity faces a Malthusian trap of widespread famine on a scale that we have not yet experienced. The geopolitical impacts of such disruptions will be severe, as an increasing number of states fail to provide their citizens with a sufficient food supply.”

What is going on here? Is this really “the biggest problem we’ve never heard of,” as Elser puts it? Or are phosphate shortages something that global markets can cope with? Let’s take a closer look.

Why we need phosphates and why we are trouble if they run out

The element phosphorus is as essential to life as carbon or oxygen. It forms part of the structure of cell walls and DNA without which no plant or animal can exist. Phosphates are phosphorus in chemical forms that are available to plants. Some phosphates occur naturally in the soil as the result of weathering of rocks, but since the dawn of agriculture, farmers have added phosphate fertilizers to increase crop production. Manure, the traditional source, still accounts for about 15 percent of all phosphates used in agriculture, but since mid- twentieth century, most such fertilizer has come from phosphate rock.

What we appear to be running out of are deposits of phosphate rock that can be mined at reasonable cost with today’s technology. Up to now, the United States has been a big producer, but its reserves are declining. China has a lot, but its domestic use is soaring and it is not a big exporter. North Africa has the biggest reserves, but some of them are in politically unstable regions like the Western Sahara.

The following widely reproduced diagram from a 2009 paper in Global Environmental Change depicts the peak phosphorus hypothesis in the form of a “Hubbert curve” that shows production declining at an accelerating rate after hitting a maximum around 2035. After that, say peak phosphate proponents, we are in big trouble.

Can the market save us?

Yes, a shortage of phosphates could spell trouble, but don’t forget about markets. Adjusting to shortages is just what markets are for. As economists see it, depleting a resource like phosphate rock is supposed to cause its price to rise. As the price rises, two things are supposed to happen. First, users are supposed to figure out ways to get by with less, and second, producers are supposed to find new sources of supply. Will this happen in the case of phosphates, or do they have unique properties that will prevent markets from working their magic?

Some think the latter. For example, the authors of the peak phosphorus diagram write that

a key difference between peak oil and peak phosphorus, is that oil can be replaced with other forms of energy once it becomes too scarce. But there is no substitute for phosphorus in food production. It cannot be produced or synthesized in a laboratory. Quite simply, without phosphorus, we cannot produce food.

Fortunately, the biological impossibility of substituting some other element for phosphorus in food production is not enough to thwart the operation of supply and demand in the phosphate market. One sign that the market is working is that phosphate prices are already rising. As the following chart shows, the U.S. prices of two of the most commonly used phosphate fertilizers soared in the early 2000s. Along with the prices of many other commodities, they dropped back from their peaks after the global financial crisis, but they are heading up again as the economy recovers.

The price increases have already had an impact on phosphate use. As the next chart shows, despite rising farm output, the growth rate of phosphate fertilizer use has slowed over time. The question for the future is whether it is technically feasible to increase food output further while actually reducing phosphate use.

Experts appear to think the answer is yes. A report published in Environmental Research Letters estimates that improvements in farm management practices and consumer waste could cut the phosphates needed to produce the present U.S. farm output by half, even with today’s technologies. In the future, even greater reductions may be possible. According to Roberto Gaxiola of Arizona State University, generations of phosphate fertilizer use have reduced the efficiency of phosphorus uptake by domesticated crop plants. His experiments indicate that selective breeding and genetic engineering can produce plants that can flourish with much lower phosphorus use.

There are significant developments on the supply side, as well. Michael Mew of the Fertecon Research Center notes that producers are already learning how to upgrade lower quality phosphate rock reserves and are modifying processing plants to accept lower quality inputs. Also, he notes that increasing vertical integration of the industry has resulted in a reduction in transportation costs. Those cost savings slow the rate of price increase and give more time for supply and demand to adjust.

Furthermore, although it is true that we cannot create or synthesize phosphorus, we can recover useable phosphorus from waste streams, including urban sewage. As this source explains, existing systems already remove phosphorus from sewage in order to preserve water quality in the rivers and streams into which they discharge treated waste. Given the low prices for phosphate that prevailed until recently, it did not pay to recover that phosphorus in usable forms. Much of it has ended up as sludge buried in landfills. However, several methods could recover a high percentage of the phosphorus from wastewater. At some price, doing so will become a profitable alternative to producing phosphate fertilizers from increasingly low-grade phosphate rock. It may even become worthwhile to mine phosphate from sewage sludge buried in old landfills.

The bottom line

The problems posed by depletion of finite supplies of high-grade phosphate rock are not trivial. However, it is highly misleading to forecast a sharp peak of phosphate fertilizer production in the near future, let alone to predict that mass starvation and population collapse lie on the downslope of the curve. The fact that there are no substitutes for phosphorus when it comes to building DNA or cell walls does not mean that markets are incapable of managing increasing scarcity.

What does seem likely is a period of continued high or rising phosphate prices, which will trigger three reactions. First, higher prices will make it economical to process ever-lower grades of phosphate rock. Second, they will spur changes in farm management and development of improved crop varieties; these in turn will accelerate incipient trends toward increased food output per unit of phosphate input. Third, higher prices will provide incentives for improved recycling of phosphorus from waste streams.

Putting all this together, Michael Mew dismisses the peak phosphate hypothesis. Instead, he foresees a phosphate plateau as higher prices cause historical growth rates to level off gradually.

Such a phosphate plateau does not preclude the need for changes in how people live and eat. It could well mean the relative price of food will rise over time, something that could cause hardship for many of the world’s poor. Furthermore, the price of phosphorus-intensive meat is likely to rise relative to those of other foods, making it unrealistic for the world’s emergent middle classes ever to attain the kind of meat-rich diet to which residents of today’s wealthy countries have become accustomed—a diet that, in the age of obesity,  is sometimes less of a blessing than a curse.

When all is said and done, a plateau is not a cliff. There is no phosphate doomsday on the horizon.

 Follow this link to view or download a slideshow version of this post, suitable for classroom use.




14 Responses to “Does Peak Phosphate Spell Doom for Humanity, or Will the Market Save Us?”

Bryan WillmanJuly 22nd, 2013 at 12:22 pm

Well argued. And a fine example of a key point – since oganisms don't convert phosphates to lead by nuclear processes, the total amount in the environment will be pretty much fixed. It's all about how hard it is to extract it and process it into a usable form.

The same applies to water – even badly contaminated water – since the contamination processes don't "destroy" or "use up" hydrogen or oxygen. But of course fully purifying contaminated salt water into a wholesome fresh water is a difficult and high energy task.

Will the gaps in technology and energy cause widespread social distress (what Wouter de Heij suggests?) I don't know. I wouldn't count it out….

TomJuly 23rd, 2013 at 9:47 am

I'm somewhere in the middle on this one. The "phosphate plateau" seems far more credible than simple Hubbard curve "peak phosphate". Just as for oil, which is already in the plateau stage. But as we've seen with oil, there can be serious stress associated with a plateau. The plateau's comforting flat line doesn't show you that in order to maintain flat output, real inputs are rising. Which amounts to a kind of reduced output. Many people are earning right at the edge of what they need to pay for food.

Indeed this one of the grimmest aspects of economics. It's nice to imagine that free markets will cope with limited resources through a smooth, calm slowdown of population growth as people are financially motivated to have less kids. I'm afraid a realistic view is that occasional waves of starvation in poorer areas will inevitably be part of the process.

TomJuly 23rd, 2013 at 9:52 am

PS I also think that, for oil and rock phosphate, the plateau eventually someday does turn into a downward slide. You can only hope on behalf of our descendants that they come up with genuine alternatives.

Ed Dolan EdDolanJuly 23rd, 2013 at 10:23 am

I agree with you that a plateau in production does not necessarily mean a plateau in price. As I said, " It could well mean the relative price of food will rise over time, something that could cause hardship for many of the world’s poor."

Meanwhile, since I wrote this, I ran across an interesting new source that documents the importance of diet to phosphate demand. It says that 72% of the increase in phosphate demand is due to increased meat consumption. (See ). To me, that underlies the idea that at least part of the market reaction to increased P price will be increased relative price of meat and slowing of the trend toward a meat-intensive diet, rather than decrease in total food consumption. Of course, that trend need not be uniform across countries or income groups.

David W. RothschildJuly 23rd, 2013 at 9:50 am

Okay, I understand the argument and want to accept its soundness. However, please
explain why neither the producers nor the investment community, especially the retail and institutional investors, seem to share this Doomsday scenario ? The share prices of producers are at disappointing lows and their yields are also paltry. Moreover, production is being scaled back because of "soft" demand in emerging markets.
In so far as the dire predictions of a collapse in population, I shall leave it to others to weigh in on that prediction but I think that Mark Twain said it best when he declared the the rumors of his death were greatly exaggerated.

Ed Dolan EdDolanJuly 23rd, 2013 at 10:28 am

Evidently, people in the investment community (Grantham aside) have read both the forecasts produced by the peak phosphate proponents and those produced by others like Mew who think both supply and demand are more elastic, and so far, they are not believing the peak phosphate hypothesis.

BTW, even if the peak phosphate hypothesis were correct, we would have to distinguish between the prices investors put on producers of phosphates, i.e., mining and processing companies, and owners of phosphate reserves. The increased rents implied by the peak phosphate hypothesis would accrue mostly to owners of reserves, not producers. I don't know how many of the producers you discuss are also owners of the reserves they mine and process, but it is something to keep in mind.

tewJuly 23rd, 2013 at 11:52 am

Just because a commodity price is increasing, doesn't mean the producers' profits increase. For mining companies it's all about reserve replacement cost. That's the key concept. If the costs of acquiring and extracting new reserves rises faster than the commodity price, then profits get squeezed.

As for "soft" demand, that's a short-term indicator. It's short-term demand vs. short-term supply conditions and tells us nothing of the longer term conditions.

tewJuly 23rd, 2013 at 11:57 am

Here's a simple way to frame oil vs. phosphate:
Oil is substitutable but not recyclable.
Phosphate is not substitutable but is recyclable.

We will ultimately stop using oil. We will always need phosphate.

Oil is a (mixture of) compounds that must be broken down to extract the energy. The value of the oil is in the compounds' chemical bonds, which are broken during use. What remains is carbon dioxide and water.

Phosphate is a simple compound and the value is in the elemental phosphorous, which remains unaltered after its use.

Chris AAugust 10th, 2013 at 7:54 am

Cullen Hendrix, in a 2011 working paper for the Peterson Institute, did a well-argued critique of the Hubbert model for phosphate forecasting. It is available at

Often, an inappropriate model or their exciting conclusions take precedence over a more scientific approach (as with Hendrix) or even a 'Hey, wait a minute here' set of ideas as posed by Dolan in this article.

James FisherAugust 13th, 2013 at 10:24 am

I used to work in phosphate exploration and there are large reserves of lower grade phosphate rock in South Carolina, Georgia and Florida in the basal Miocene section. some of these phosphate deposits also contain enough uranium to be classified as a uranium ore as well as a phosphate source. Obviously the key factor is the cost of recovering the phosphate. In North Carolina open pit mines can recover phosphate rock for less than $50 per ton in an open pit mine. The phosphate beds there are about 40 feet thick containing 20-40k tons phosphate per acre. South Carolina and Georgia have extensive reserves of phosphate in beds 20 feet thick, and containing 5-15k tons of phosphate per acre. There are also marine reserves of phosphate rock off the coast of Northern South Carolina. Open pit mining in these sediments creates a huge hole which can be reclaimed to nearly normal topography for the area. During mining the overburden expands and will more than fill the hole after removal of the phosphate. The key questions that need to be answered are the environmental costs of disturbing the land. A large part of the land in South Carolina is swamp land with highly acidic water that is toxic enough that mosquitos do not live there. About half of the area in South Carolina and Georgia is highly productive farm land. Taking some of this land out of production and mining the phosphate may be of benefit to society however the cost to the environment would need to be acceptable.

tank trouble onlineOctober 13th, 2017 at 6:21 am

Play the most interesting tank trouble game forever play this online create the best the best fun to get the site most popular in the world wide i am always ready to get the site so have the most enjoy with your friends thanks for the site.