More on China's Economy, Food Production, and Food Demand

Over the next two decades the Panda may begin to feel peckish. A hard look at China's food production and resource availability suggests more difficult times ahead. And this is just one potential problem. Throughout my travels and reading over the past 5 years, I have noticed that people with lots of experience on the ground in China question whether the current pace of development is sustainable.

The upshot of all this may be that the easy gains have been made. In the years to come, China will be faced with extremely hard choices about how to simultaneously maintain economic growth, clean up its environment, and feed its population, particularly when it appears that most of the expected increase in food demand due to rising incomes has yet to be realized. So, following up on last week's post about The Future of China's Economy, here are a few more thoughts that frame future potential stumbling blocks.

Running Out of Cheap Labor, and Coming Home for L.A.'s "Clean" Air

John Pomfret, formerly the Beijing Bureau Chief for the Washington Post, definitely has a lot of experience in country. In "A Long Wait at the Gate to Greatness", he asks, "Is China really going to be another superpower?"

His short answer is, "I doubt it." In more depth:

It's not that I'm a China-basher, like those who predict its collapse because they despise its system and assume that it will go the way of the Soviet Union. I first went to China in 1980 as a student, and I've followed its remarkable transformation over the past 28 years. I met my wife there and call it a second home. I'm hardly expecting China to implode. But its dream of dominating the century isn't going to become a reality anytime soon.

Too many constraints are built into the country's social, economic and political systems. For four big reasons -- dire demographics, an overrated economy, an environment under siege and an ideology that doesn't travel well -- China is more likely to remain the muscle-bound adolescent of the international system than to become the master of the world.

Pomfret goes through the same sort of list of potential stumbles that I compiled for last week's post, and adds a few more. He notes that that population control policy has produced an inverted population pyramid, which requires a smaller, young population cohort to support a larger, older cohort as the latter leave the workforce. This while life expectancy has more than doubled in the last fifty years. Thus the expectation is that the workforce will shrink over the coming decades, labor costs will rise, and more of that labor will be put toward supporting non-working elders.

Pomfret also observes that:

One important nuance we keep forgetting is the sheer size of China's population: about 1.3 billion, more than four times that of the United States. China should have a big economy. But on a per capita basis, the country isn't a dragon; it's a medium-size lizard, sitting in 109th place on the International Monetary Fund's World Economic Outlook Database, squarely between Swaziland and Morocco. China's economy is large, but its average living standard is low, and it will stay that way for a very long time, even assuming that the economy continues to grow at impressive rates.

Unlike many observers, he doesn't discount the potential drag on economic growth from pollution, leading off with a personal anecdote:

When my family and I left China in 2004, we moved to Los Angeles, the smog capital of the United States. No sooner had we set foot in southern California than my son's asthma attacks and chronic chest infections -- so worryingly frequent in Beijing -- stopped. When people asked me why we'd moved to L.A., I started joking, "For the air."

Pomfret is perplexed about why Westerners seem to be ignoring pollution's ~10% hit to the Chinese GDP: "Somehow, though, the effect this calamity is having on China's rise doesn't quite register in the West." As I discussed in the earlier post, this shortsightedness confuses me, too, particularly when you combine the effects of pollution with the demands on domestic water and land to provide food for a hungry population.

Missing Food Demand

In a report last year from the Economic Research Service of the USDA, Fred Gale and Kuo Huang suggest that China may face increasing difficulties in meeting domestic food demand. I find their argument quite compelling and will later state it even more firmly than they do.

Gale and Huang observe that growth in food demand has, unexpectedly, not kept pace with overall economic growth. Here is the conundrum: "Given the responsiveness of food demand to income growth, China's rapid growth of 9-10 percent per year suggests that its demand for food is growing faster than its production capacity. ...How is it that China's surging income growth has not pushed its demand for food beyond its domestic production capacity?"

The main factor the authors identify is that while a small, wealthy fractionof the population now evidently has enough to eat, and thus spends additional income on quality rather than quantity, a large majority of consumers have yet to fill their bellies.

The underlying cause for lagging food demand is not surprising once you think about it. Because economic benefits, in particular income gains, disproportionately go those with already high incomes, and because those with high incomes tend to spend on quality rather than quantity, the total volume of food consumed by the Chinese population has risen only slowly. The authors note that:

...Expenditures by the top tier of households--China's emerging class of professionals and entrepreneurs -- have grown at double-digit rates. Food expenditures were nearly stagnant for the bottom 20 percent of urban households. Food expenditures by rural households grew 2.6 percent annually.

...Income growth for low-income urban and rural households--the majority of China's households--was well below GDP growth. ...Average income for the lowest decile of urban households actually declined slightly between 2000 and 2003.

This suggests to Gale and Huang that, "Food consumption and income growth patterns may explain how China has been able to remain self-sufficient in most food items." The authors stop their argument here, but I think they could go further.

The Still-Hungry ~1 Billion

The lag between GDP growth and food consumption has important implications for future increases in food demand.

Based on the statistics compiled by Gale and Huang, it looks to me like more than 90% of the Chinese population has a per capita annual income below 10,000 Yuan. This is an interesting figure for considering future food demand because Gale and Huang also demonstrate that pork consumption in China continues to rise as a function of income until about 10,000 Yuan. Poultry and seafood consumption also rise strongly as a function of income, but notably don't saturate like pork at 10,000 Yuan. More meat consumption requires more grain and more water to raise the animals (see a previous post, "China and Future Resource Demands").

Here is where I think the argument could be made more forcefully.  As best I can make out, what all the above means is that most of the increase in food demand we might expect from rising incomes in China has yet to be realized; more than 80% of the population is, "Still at income levels where they demand increased quantities of many foods as their income rises."

So where is China going to get all this food? One answer is imports, another is to go offshore to buy or rent farmland (see the "China and Future Resource Demands" post), and yet another is to push domestic production. But the latter may be difficult.

"Who Will China Feed?"

This is the question asked by Fred Gale and Bryan Lohmar in an essay in Amber Waves, the USDA magazine. They elaborate their surprise at China's ability to feed its population: "While China has emerged as the world's leading importer of soybeans, vegetable oil, cotton, wool, rubber, and animal hides, it has been surprisingly successful at meeting the basic food needs of its population of more than 1.3 billion people, and it has stepped up as a major food exporter."  (They make no mention of the income inequality and consequent food spending gap explored above.)

Given the pace of growth and limited resources, they ask, "How long can China sustain this momentum?"

China imports only small amounts of premium-grade rice, minor amounts of wheat in most years, and no corn. China has maintained agricultural self-sufficiency in grains as it carries out the world's largest and fastest urbanization and industrialization. Economic development is increasing competition for scarce resources in China, but growing incomes are allowing most consumers to increase consumption of fruit, vegetables, and livestock products.

China has become a significant food exporter by ramping up production in many sectors and gaining world market share. Indeed, China has been a net food exporter for most of the last three decades. China dominates world markets in a variety of products areas, including garlic, apples, apple juice, mandarin oranges, farm-raised fish and shrimp, and vegetables. At times, it seems that China has suspended the law of scarcity by boosting production in many sectors and selling at low prices without having to sacrifice production in other sectors.

One way to look at this is that China is exporting high value "food products", not staples that the majority of Chinese themselves consume. This strategy contributes to the trade surplus, but the use of land to grow crops for export must clearly be balanced with domestic demand for staples. This balance also points to the fact there is some room for moving crop land now used for exports back into production to satisfy domestic demand.

Here are two key paragraphs on how China has increased its food production yields:

Investments in research and development raised the quality of inputs and the efficiency of their use over the past two decades. Research into improved varieties and quality of seeds surged after the late 1970s. By the turn of the century, China had more agricultural researchers than any other country, and a larger budget for public sector agricultural research than any developing country. Fertilizer quality in China also has improved over the past two decades, as farmers move away from applying pure nitrogen fertilizer to applying more nitrogen-phosphorous- potassium blends. China has been importing breeding animals--which are often crossed with domestic breeds--to improve efficiency of weight gain, improve disease resistance, and raise milk output. The government has offered subsidies to farmers for dairy herd improvement for several years.

China today is the world's largest agricultural producer and consumer. With an estimated 10 percent of world land resources and 6 percent of world water resources, China produces 30 percent of the world's rice, 20 percent of the world's corn, a fourth of the world's cotton, an estimated 37 percent of the world's fruit and vegetables, and half of the world's pork. For most products, China's world share of production is close to or exceeds its 20-percent share of world population. China, however, has exploited the means of coaxing food and fiber out of a limited natural resource base to the extent that additional gains will be more difficult than in the past.

Gale and Lohmar go on to discuss water and soil quality issues, fertilizer and pesticide use, and industrial pollution, while briefly addressing labor costs:

China has been able to maintain low-cost production in international agricultural markets largely because of low labor costs. Historically, Chinese farms have raised large amounts of output from small plots by using labor-intensive production strategies, such as growing multiple crops per year, intercropping, and growing vegetables in courtyards. But hundreds of millions of rural workers have found nonfarm employment over the last two decades. The flow of labor from rural areas enabled China's industry and cities to boom, while wage growth was relatively stagnant for much of the last two decades.

China's rapid economic expansion appears to have finally exhausted the pool of under-employed workers. Since 2003, wages have been rising at a double-digit pace. The dwindling pool of available rural workers is resulting in increased mechanization of harvesting and planting. Anecdotal evidence also suggests that intensive agricultural practices, like double-cropping, transplanting seedlings by hand, and small-scale hog production, have decreased due to labor shortages and high wages.

So, as John Pomfret suggested in his piece in the Washington Times Post (!), labor costs are already affecting food production. But the bigger issue is in trying to identify where, exactly, future gains in production are going to come from.  Rough estimates of the probable increase in demand give some context for the magnitude of the problem.

Returning to the correlation of meat consumption and income: It appears from FAO and USDA data that China is bound to eat more meat, especially pork, as incomes continue to rise.  Growing meat for human consumption creates a big lever in water and grain markets.  Producing a kilo of pork requires approximately three kilos of grain, and producing a kilo of beef requires about eight kilos of grain.  Based on the data in Gale and Huang, in appears that as income rises from 3000 to 10,000 Yuan, pork consumption increases by about 50%, to ~23 kg, which will require about 70 kg of grain.  This in addition to the ~30% increase in grain products (~6 kg) directly purchased by households as incomes rise over that range.  Fish and poultry demand about doubles, too, from ~8 to ~16 kg per capita, but estimating the additional grain consumption here is hard.  I'll hand wave and make a low-ball estimate that it will take only another 16 kg of grain to feed the the fish and poultry.

Adding this all together, that is an additional per capita  increase in grain demand of more than 90 kg.  Here is the kicker: that number appears to hold for at least 500 million people, perhaps as many as a billion.  That amounts to at least 45 million tonnes (metric!) of grain, perhaps as much as 90 million tonnes.  The Chinese population would then still be consuming only about 80% as much animal protein per capita as Europeans, and only a little over half as much as us gluttons in the U.S.

China produces about 500 million tonnes of grain per year (see the USDA ERS China Ag and Economic data page), so supplying increased meat demand with domestic grain supplies would require a (very rough) increase of between 10 and 20% in total yield.  That doesn't necessarily sound like much -- I actually expected the increase to be a larger percentage of current harvests -- and might be accomplished by breeding, genetic modification, and better farming practices.  But as detailed in my earlier posts, China is losing both arable land and usable water.  With only 7% of the globe's arable land to work with (ignoring losses to due climate change and prior poor farming practices), the country is going to have to work very hard indeed to squeeze more grain out of those limited resources.

That leaves imports, which means competing on the world commodity markets for food.  In combination with rising labor costs at home, all this points to rising domestic prices and rougher going for the Chinese economy.

"The Pickens Plan" for Wind Energy: Why Use Natural Gas for Cars?

Oilman T. Boone Pickens made a splash last week by announcing plans to build a wind farm with 4,000 megawatts worth of generating capacity.  The Pickens Plan calls the U.S. the Saudi Arabia of oil wind, and he notes that, "At current oil prices, we will send $700 billion dollars out of the country this year alone — that's four times the annual cost of the Iraq war."  His logic in making this investment is pretty straightforward:

Building wind facilities in the corridor that stretches from the Texas panhandle to North Dakota could produce 20% of the electricity for the United States at a cost of $1 trillion. It would take another $200 billion to build the capacity to transmit that energy to cities and towns.

That's a lot of money, but it's a one-time cost. And compared to the $700 billion we spend on foreign oil every year, it's a bargain.

Great -- the more energy we generate at home, the more we can invest in rebuilding the U.S. economy and infrastructure.  Somewhat less obvious is the logic of his suggestion that this electricity be used to free up natural gas now burned to provide ~20% of US electricity, and instead use that gas to power cars.

There are very few natural gas powered cars in this country, and it would take an enormous investment to either retrofit existing vehicles or replace a large fraction of the existing fleet in less time than the present ~13 year life cycle.  Moreover, burning natural gas in large turbines is way more efficient than burning it in small car engines, so it is actually better used to produce electricity for the grid.  It would seem to make more sense to just use the added electricity generation capacity from wind to directly offset petroleum use.

Why not just replace or retrofit the fleet with plug-in hybrids that substantially increase the efficiency of cars regardless of their fuel type?  Then you could be agnostic about the specific engine technology and fuel, but still know you could potentially double the mileage of any given vehicle by recharging from the electricity grid?  Here, for example, is a story at Wired News by Chuck Squatriglia in which Andy Grove, the CEO of Intel, calls for converting 10 million cars and trucks in the U.S. to plug in hybrids over the next four years.  The story quotes John Dabels, CEO of conversion start-up EV Power Systems, as saying his company can provide an $11,000 conversion kit that bolts onto the transmission of existing cars and trucks and delivers a 30-40% increase in liquid fuel efficiency.  Google has evidently been running a fleet of plug-in Priuses and Escapes with a 50% improvement over the standard hybrid.  These are early numbers.  Efficiencies are bound to increase as better batteries and electric motors enter the market.

Pair plug-in hybrids with microbial biofuel synthesis -- oh, alright, and even cellulosic ethanol -- and suddenly you get way more out of your feedstock and thereby reduce pressure on food prices.  Not that I am biased or anything.

The Future of China's Economy

It's hot and damp in southeastern China this time of year.  So reports a relative of mine working in the area who called to chat a few days ago.  He was suffering through another day without air conditioning, in the middle of yet another regional power outage due to a shortage of coal.  This occurrence is evidently not uncommon.  We hear a great deal in the U.S. about the unstoppable juggernaut of the Chinese economy, but sometimes I wonder if the Chinese aren't setting themselves up for a stumble or two.

(Update: For more on resource demands, see my subsequent post "More on China's Economy, Food Production, and Food Demand".)

Many of the signs point to inevitable economic superiority.  The Carnegie Endowment for International Peace released a report last week that projects China's economy will overtake that of the U.S. by 2035 (Yahoo News). "China's Economic Rise--Fact and Fiction", by Albert Keidel, concludes that China's economy is now dominated by internal growth rather than exports, and that China's economy will be twice that of the U.S. by 2050.  Keidel gives the nod to financial and bureaucratic tangles as the primary threats to growth, but does not appear particularly concerned about environmental damage and pollution. He argues that:

The record for several other East Asian economies argues that pollution is unlikely to undermine China's growth in the coming decades. In particular, Japan, South Korea, and Taiwan all passed through similar periods of serious pollution associated with rapid industrialization. In these cases, policy responses were also delayed but eventually reduced pollution levels that in some dimensions were worse than China's today.

Maybe so, but, depending on how you look at the numbers, the cost of pollution may be wiping out all of China's GDP growth.

(Update 25 July, 2008: Here is a video feed from Fora.tv of a panel discussion at the Carnegie Endowment for International Peace discussing the "Fact and Fiction" report.  I haven't watched the whole thing yet...)

Attempting to Account for the Costs of Pollution

For most of the last decade, China's government has downplayed the cost of environmental damage to the country's GDP.  However, according The Economist, in March of 2008, Pan Yue, a deputy minster at the State Environmental Protection Agency (SEPA), publicly estimated that environmental damage reduces GDP by as much as 13%.  As recently as May, 2006, the official estimate was only 3% of GDP for 2004, a tally contained in the first and only "green audit" of the economy.

The direct costs to human life are substantial, but official estimates are also variable.  A study by SEPA and The World Bank, published last year, "The Cost of Pollution in China", estimates that pollution is directly responsible for at least 750,000 deaths a year, while in a 2006 speech Mr. Pan stated that approximately 70% of China's two million annual cancer deaths were caused by pollution.

The disparity in these figures is evidently caused by political tension between different parts of the Chinese government.  Both the health findings and the future of the "green GDP audit" were evidently compromised by political infighting between state scientists, regional leaders, and officials in other ministries.  The New York Times reported that:

The official explanation was that the science behind the green index was immature. Wang Jinnan, the leading academic researcher on the Green G.D.P. team, said provincial leaders killed the project. "Officials do not like to be lined up and told how they are not meeting the leadership's goals," he said. "They found it difficult to accept this."

Here is the point: Even a 10% reduction in Chinese GDP would, in effect, zero out the overall growth of the economy.  Viewed this way, despite its role in the global economy, any "wealth creation" and growth in China may be accounted for entirely by the cost of degrading the local environment and increasing human disease and death.  You can understand how government officials might be uneasy about publicizing this figure.

According to officials at The World Bank, its "Cost of Pollution" report was similarly abridged for political reasons; "China's environmental agency insisted that the health statistics be removed from the published version of the report, citing the possible impact on 'social stability'."  As a result, one-third of the document was reportedly withheld from publication.  The tension between open communication and central control, and between development and damage, is evident in a press release from Gov.cn, the Government's official web site:

Even though the economic growth characterized by "high consumption, high pollution and high risk" is of its own historical significance in China, China's economy has been in the bottleneck period of resources and energy today and it cannot bear any risks of resources exhaustion.

Meanwhile, Chinese society has also entered the period with various conflicts protruding in which per capita GDP is about 1,000-3,000 US dollars, which cannot bear up any social problems caused by environmental pollution.

The government is clearly aware of the social and economic threats of environmental damage.  As reported by the Shanghai Daily, the most recent five year plan; "Requires energy consumption per unit of GDP to decline by 20 percent from the previous planning period.  The total amount of major pollutants discharged will be reduced by 10 percent, and forest coverage will be raised from 18.2 percent to 20 percent."

In an effort better address environmental concerns, in March of 2008 the State Council upgraded SEPA to a full Cabinet-level ministry.  To gather, "Accurate and high-quality data [of] pollution sources," the government launched in the first pollution census in February 2008.  And yet even while the central government attempts to close illegal and polluting coal mines and coal burning plants, journalists regularly report that local and regional authorities either ignore or explicitly condone the reopening of those facilities (1, 2, 3).

It does not appear that China's reliance on coal is going to decrease any time soon.  China has recently been building coal-fired plants at the rate of one every 7 to 10 days, with plans to build 500 more over the next decade.  The fraction of newly built power plants that burn coal has increased from 70% to 90% since 2000.  Thus, without either a more unified approach to reducing pollution or a substantially stronger response to that end by the central government, environmental damage will continue to directly plague both the economy and human health.

The Future Cost of the Building Boom

Here is something I don't see discussed in the press: where are the Chinese going to get all the coal to fire all the new power plants, especially when they are already facing supply shortfalls?  (Update: To clarify, I am less concerned here with the amount of coal in the ground than supply chain issues.  If they are already having trouble moving coal quickly enough to existing plants, how will they manage the increased demand?)  And while they may have the coal in the ground, how much will it cost to mine it with labor costs rising all across the country? And what about the costs of additional transportation infrastructure?

This brings us back to my father-in-law, sweltering away in Xiamen, and one of stumbling blocks the Chinese may be literally building for themselves.  As reported by The New York Times:

Each year for the past few years, China has built about 7.5 billion square feet of commercial and residential space, more than the combined floor space of all the malls and strip malls in the United States, according to data collected by the United States Energy Information Administration.

Chinese buildings rarely have thermal insulation. They require, on average, twice as much energy to heat and cool as those in similar climates in the United States and Europe, according to the World Bank. A vast majority of new buildings -- 95 percent, the bank says -- do not meet China's own codes for energy efficiency.

All these new buildings require China to build power plants, which it has been doing prodigiously. In 2005 alone, China added 66 gigawatts of electricity to its power grid, about as much power as Britain generates in a year. Last year, it added an additional 102 gigawatts, as much as France.

Damn.  So not only is China building enormous power generation capacity, but their underlying infrastructure is inherently inefficient.  This kind of systemic inefficiency is often attributed to, excused, or even just written off as a characteristic of a particular "stage of economic development" (see Keidel for one example).  It is certainly true that the U.S., Japan, and Europe all went through periods when the focus was on generating jobs and building wealth, only later to be followed up by mining inefficiencies to squeeze more product out of each unit of water and energy.

But all of China's infrastructure, all that housing and commercial space, is brand new.  So here are more questions: Is the government's plan to replace inefficient buildings over the next couple of decades?  Will labor remain so inexpensive that Chinese infrastructure is, in effect, disposable?  What are the secondary costs of maintaining that construction boom (e.g., energy, pollution, materials)?

It would seem that without a truly radical change in energy production, China is setting itself up to rely on dirty coal for many decades to come.  And so I wonder: How it is that the country will escape continued environmental damage that is equivalent to China's GDP growth?  You can't put off dealing with those costs forever.

The U.S. is borrowing cash from China to buy petroleum, and we have to sort that out as soon as possible.  But the Chinese are using their health, and thus their future productivity, as collateral for present growth. Even if they don't stumble, they may have to pause to catch their breath.

Ineffective Export Controls for US Technology

In the context of my ongoing skepticism about the effectiveness of regulation for improving biosecurity, here is a quick note on the utility of export controls for restricting transfer of sensitive technology.

Over at Wired News, Noah Shachtman has a post pointing to an article in Mother Jones about all the US weapons that are winding up in the hands of Iran.  Re-sale by third parties seems to be the short answer, but read the article to get the full story.

DNA Synthesis "Learning Curve": Thoughts on the Future of Building Genes and Organisms

With experience comes skill and efficiency.  That is the theory behind "learning" or "experience curves", which I played around with last week for DNA sequencing.  As promised, here are a few thoughts on the future of DNA synthesis.  Playing around with the synthesis curves a bit seems to kick out a couple of quantitative metrics for technological change.

For everything below, clicking on a Figure launches a pop-up with a full sized .jpg.  The data come from my papers, the Bio-era "Genome Synthesis and Design Futures" report, and a couple of my blog posts over the last year.

Figure 1.

The simplest application of a learning curve to DNA synthesis is to compare productivity with cost.  Figure 1 shows those curves for both oligo synthesis and gene synthesis (click on the figure for a larger pop-up).  These lines are generated by taking the ratios of fits to data (shown in the inset).  This is necessary due to the methodological annoyance that productivity and cost data do not overlap -- the fits allow comparison of trends even when data is missing from one set or another.  As before, 1) I am not really thrilled to rely on power law fits to a small number of points, and 2) the projections (dashed lines) are really just for the sake of asking "what if?".
 

What can we learn from the figure?  First, the two lines cover different periods of time.  Thus it isn't completely kosher to compare them directly.  But with that in mind, we come to the second point: even the simple cost data in the inset makes clear that the commercial cost of synthetic genes is rapidly approaching the cost of the constituent single-stranded oligos. This is the result of competition, and is almost certainly due to new technologies introduced by those competitors.

Assuming that commercial gene foundries are making money, the "Assembly Cost" is probably falling because of increased automation and other gains in efficiency.  But it can't fall to zero, and there will (probably?) always be some profit margin for genes over oligos.  I am not going to guess at how low the Assembly Cost can fall, and the projections are drawn in by hand just for illustration.

Figure 2.

It isn't clear that a couple of straight lines in Figure 1 teach us much about the future, except in pondering the shrinking margins of gene foundries.  But combining the productivity information with my "Longest Synthetic DNA" plot gives a little more to chew on.  Figure 2 is a ratio of a curve fitted to the longest published synthetic DNA (sDNA) to the productivity curve.

In what follows, remember that the green line is based on data.

First, the caveat: the fit to the longest sDNA is basically a hand hack.  On a semilog plot I fit a curve consisting of a logarithm and a power law (not shown).  That means the actual functional form (on the original data) is a linear term plus a super power law in which the exponent increases with time.  There isn't any rationale for this function other than it fits the crazy data (in the inset), and I would be oh-so-wary of inferring anything deep from it.  Perhaps one could make the somewhat trivial observation that for a long time synthesizing DNA was hard (the linear regime), and then we entered a period when it has become progressively easier (the super power law).  I should probably win a prize for that.  No?  A lollipop?

There are a couple of interesting things about this curve, along which distance represents "progress".  First, so far as I am aware, commercial oligo synthesis started in 1992 and commercial gene foundries starting showing up in 1999.  The distance along the curve in those seven years is quite short, while the distance over the next nine years to the Venter Institute's recent synthetic chromosome is substantially larger.

This change in distance/speed represents some sort of quantitative measure of accelerating progress in synthesizing genomes, though frankly I am not yet settled on what the proper metric should be.  That is, how exactly should one measure distance or speed along this curve?  And then, given proper caution about the utility of the underlying fits to data, how seriously should one trust the metric?  Maybe it is just fine as is.  I am still pondering this.

Next, while the "learning curve" is presently "concave up", it really ought to turn over and level off sometime soon.  As I argued in the post on the Venter Institute's fine technical achievement, they are already well beyond what will be economically interesting for the foreseeable future, which is probably only 10-50 kilobases (kB).  It isn't at all clear that assembling sDNA larger than 100 kB will be anything more than an academic demonstration.  The red octagon (hint!) is positioned at about 100 MB, which is in the range of a human chromosome.  Even assembling something that large, and then using it to fabricate an artificial human chromosome, is probably not technologically that useful.  I reserve a bit of judgement here in the event it turns out that actually building functioning human chromosomes from smaller pieces is problematic.  But really, why bother otherwise?

Figure 3.

Next, with the other curves in hand I couldn't help but compare the longest sDNA to gene assembly cost (beware the products of actual free time!).  (Update: Can't recall what I meant by this next sentence, so I struck it out.) Figure 3 may only be interesting because of what it doesn't show.  Note the reversed axis -- cost decreases to the right.

The assembly cost (inset) was generated simply by subtracting the oligo cost curve from the gene cost curve (see Figure 1 above) -- yes, I ignored the fact that those data are over different time periods.  There is no cost information available for any of the longest sDNA data, which all come from academic papers.  But the fact that gene assembly cost has been consistently halving every 18 months or so just serves to emphasize that the "acceleration" in the ratio of sDNA to assembly cost results from real improvements in processes and automation used to fabricate long sDNA.  I don't know that this is that deep an observation, but it does go some way towards providing additional quantitative estimates of progress in developing biological technologies.

"Learning Curves" and Genomics: Thoughts on the Future of Sequencing

(Update: 23 March 2009, I fixed various broken links.)

I have been wondering what additional information about future technology and markets can be discerned from trends in genome synthesis and sequencing ("Carlson Curves").  To see if there is anything there, I have been playing around with applying the idea of "learning curves" (also called "experience curves") to data on cost and productivity.

Learning curves generally are used to estimate decreases in costs that result from efficiencies that come from increases in production.  The more you make of something, the more efficient you become.  T.P. Wright famously used this idea in the 1930s to project decreases in cost as a function of increased airplane production.  The effect also shows up in a reduction of the cost of photovoltaic power as a function of cumulative production (see this figure, for example).

To start with here are some musings about the future of sequencing and the thousand dollar genome:

Figure 1 was generated using data on sequencing cost and productivity using commercially available instruments (click on the image for a larger pop-up).  I am not yet sure how seriously to take the plot, but it is interesting to think about the implications.

A few words on methodology: the data is sparse (see inset) in that there are not many points and data is not readily available in each category for each year.  This makes generating the plot of cost vs. productivity subject to estimation and some guesswork.  In particular, fitting a power law to the five productivity points, which are spread over only three logs, makes me uneasy.  The cost data isn't much better.  In the past I have cautioned both the private sector and governments from attempting to use this data to forecast trends.  But, really, everyone else is doing it, so why should I let good sense stop me?

Before going on, I should note that sequencing cost and productivity are related but not strictly correlated.  They are mostly independent variables at this point in time.  Reagents account for a substantial fraction of current sequencing costs, and increasing throughput and automation do not necessarily affect anything other than the number of bases one person can sequence in a day.  It is also important to point out that I am plotting productivity rather than cumulative production, and that both productivity and cost improvements include changes to new technology.  So the learning curve here is sort of an average over different technologies.  It is not a standard way to look at things, but it allows for a few interesting insights.

The blue line was generated by taking a ratio of fits to both the cost and productivity lines.  In other words, the blue line is basically data, and it suggests that for every order of magnitude improvement in productivity you get roughly a one and a half order of magnitude reduction in cost.  Here is the next point that makes me uneasy: I really have no reason to expect the current trends to maintain their present rates.  New sequencing technologies may well cause both productivity and cost changes to accelerate (though I would not expect them to slow -- see, for example, my previous post "The Thousand Dollar Genome").

Forging ahead, extending the trend out to the day when technology provides for the still-mythical Thousand Dollar Genome (TGD) provides an interesting insight.  At present rates, the TGD comes when an instrument allows for a productivity of one human genome per person-day.  It didn't have to be that way; slightly different doubling times (slopes) in the fits to cost and productivity would have produced a different result.  Frankly, I don't know if it means anything at all, but it did make me sit up and look more closely at the plot.  You could even call it a weak prediction about technological change -- weak because any deviation from the present average doubling rates would break the prediction.

But even if the present rates remain steady, that doesn't mean the actual cost of sequencing to the end user falls as quickly as it has.  Let's say somebody commercially produces an instrument that can actually provide a productivity of one genome per person-day.  How many of those instruments might make it onto the market?

Let's estimate that one percent of the US population wants to sign up for sequencing.  Those three million people would then require three million person-days worth of effort to sequence.  Operating 24/7 for one year, that would require just over 2700 instruments.  It will take some time before that many sequencers are available, which means that even if the technological capability exists there will be some -- probably substantial -- scarcity (the green circle on Figure1 ) keeping prices higher for some period.  Given that demand will certainly extend into Europe and Asia, further elevating prices, there is no reason to think the TGD will be a practical reality until there exists competition among providers.  This competition, in turn, will probably only emerge with the development of a diverse set of technologies capable of hitting the appropriate productivity threshold.

What does this imply for the sequencing market, and in particular for health care based on full genome sequencing?  First, costs will stay high until there are a large number of instruments in operation, and probably until there are many different technologies available.  Thus, if prices are determined solely by the market, the idea of sequencing newborns to give them a head start on maximizing their state of health will probably be out of reach for many years after the initial instrument is developed.  Market pricing probably means that sequencing will remain a tool of the wealthy for many, many years to come.

So, what other foolish, over-extended observations can I make based on fitting power laws to sparse data?  Just one more for the moment, and it actually doesn't depend so much on the actual data.  At a productivity of one genome per person-day, you really have to start thinking about the cost of that person.  Somebody will be running the machine, and that person draws a salary.  Let's say that this person earns a technician's wage, which amounts with benefits to $300/day.  All of a sudden (the trends are power laws, after all) that is 30% of the $1000 spent on sequencing the genome.  If the margin is 10-20% of the cost, then the actual sequencing, including financial loads such as depreciation of the instrument and interest, can cost only $500.  We are definitely a long time from seeing that price point.

I'll post on the learning curve for genome synthesis after I make more sense of it.

"The Big Squeeze: New Fundmantals for Food and Fuel Markets"

Bio-era recently released a new report describing our latest thinking about the future of food and fuel markets.  In the short term, we could be in for an even bumpier ride than we have seen so far.  Over the longer term, new technologies (biological and otherwise) will profoundly alter our ability to produce non-fossil fuels and will thus alter the structure of the economy.  But the sheer size of the petroleum and gasoline markets will continue dominate energy markets, and our economy, for many years to come.

Click on the image to  obtain the report -- as with previous releases you can purchase a copy from a print on demand service or download a PDF after registering.

Here is the Introduction:

In recent years, rising prices for agricultural and energy commodities have heightened interest in the economic fundamentals governing these markets. This report presents bio-era’s latest thinking on some of these fundamentals, and how they may be changing in unanticipated ways. Part of what we explore here concerns the interactions between the principal “long forces” affecting these markets, including the forces of climate change, the limits of conventional crude oil supply expansion, and the impacts of continued underlying growth in global populations and economies. Not surprisingly, we foresee these long forces acting in combination to place additional upward pressure on fuel and food prices, and we present a model for thinking about the dynamics at work in what we hope is a simple, but useful, way.

In addition, we also consider the growing linkages between agricultural and energy commodities, and how these linkages might affect current and future pricing dynamics within and between these markets. Under one, very specific set of conditions, we believe that price signaling between these markets could lead to a self-reinforcing feedback loop — which if left unchecked — could result in steadily escalating clearing prices.  The theoretical effect we describe is akin to an “evolutionary arms race” or a “red queen effect.” Should market circumstances ever give rise to the price dynamic described here, the implications could be far-reaching. Energy and food prices could rise steadily as a result, at great cost to the global economy. Continuing globalization might even be placed at risk. For these reasons, and because these theoretical possibilities have gone largely unnoticed to date, we felt it worth calling special attention to them here.

Here are the "Key Findings":

• Despite seven years of rising real prices for crude oil and a doubling of prices over the past year, global crude oil production has been nearly flat since 2005.
• The production of biofuels--in the form of ethanol fermented from sugars and starches, and biodiesel derived from vegetable oils and animal fats - has increased significantly and is now an important source of supply satisfying year over year increases in global liquid transportation fuel consumption.
• There are two principal connections between the crude oil and petroleum product markets and many of the so-called "soft" agricultural commodities such as grains, sugar, and vegetable oils:

1. an input-cost effect on agricultural commodity prices because oil and energy-intensive fertilizers account for a significant share of total production costs for most major crops;
2. an output-price effect prices of petroleum products such as gasoline or diesel oil set a floor price for agricultural commodities that can be converted into fuel substitutes.

• The first of these connections--the input-cost effect--is "one-way." The cost of petroleum will influence agricultural commodity prices over time, but the reverse is not true--the cost of agricultural commodities will have little or no effect on the costs of producing, transporting, and refining petroleum.
•  The second of these connections--the output-price effect--is increasingly "two-way." As volumes of agriculturally-derived fuels grow, expanding or withholding these volumes from the petroleum product markets directly influences both the price of petroleum products and the price of agricultural commodities.
• The result is competition between food and fuel end-use markets to price at a level sufficient to attract (and/or preserve access to) marginal supplies. Attempting to hold down food prices by restricting or redirecting feedstocks used to produce fuel, may cause fuel prices to rise. Similarly, attempting to hold down or lower fuel prices by increasing conventional biofuels production may increase food prices.

In the absence of a supply response from conventional crude oil, looking ahead, this dynamic is expected to continue until either global economic growth slows substantially, or additional supplies of non-conventional fuel substitutes - such as gas-to-liquids, coal-to-liquids, or biomass-to-liquids -- become available at meaningful scale. The necessary lead time on the latter option is at least 3-5 years.

"The Big Squeeze: New Fundamentals for Food and Fuel Markets",  A Special Bio-era Report, June 2008, By Stephen C. Aldrich, James Newcomb, Dr. Robert Carlson

More Pieces in the Distributed Biofuel Production Puzzle

Here are some additional musings on distributed production of biofuels and economies of scale:

Following on last month's launch of the efuel100 Microfueler, which seems to be a step toward distributed biofuel production, comes word of a couple of high school students who built a "Personal Automated Ethanol Fermenter and Distiller" (via Wired) for the 2008 Intel International Science and Engineering Fair.

In the video, Eric Hodenfield and Devin Bezdicek don't give a great deal of detail about their project, but I think it is fascinating that a couple of high school students decided to build a widget intended to facilitate personal fuel production.  Kudos to those two.  The device, like the Microfueler, is supposed to produce ethanol on a small scale, but both would be useful to produce Butanol instead if the appropriate microbe were handy, as I have written about before.

But why stop there?  What about home production of petroleum?  The TimesOnline this week has a short story about LS9, featuring Greg Pal, who suggests the company has a microbe with the capability to produce petroleum at $50 per barrel using Brazilian sugar as a feedstock.  (See my earlier post LS9 - "The Renewable Petroleum Company" - in the News.)  That number is interesting, because when I met Mr. Pal last fall at a retreat organized by Bio-era, he was more reticent about proposing a target price.  It would seem that the company is making decent progress, with Mr. Pal suggesting to the Times that LS9 hopes to be producing fuel on a commercial scale by 2011.

The Times article goes on to list some rather large sounding figures for the land that might be required to supply the US fuel weekly demand of ~140 million barrels using microbes; "205 square miles, an area roughly the size of Chicago".  Skipping the issue of whether there is enough sugar produced around the world to use as feedstock, the choice of paving Chicago over to crank out a weekly supply of renewable petroleum is a little odd.  Simplifying the calculation makes the whole problem seem quite reasonable.

First, consider that US daily oil consumption is something like 20 million barrels, according to the DOE.  So, if in practice biofuel production is no more efficient than LS9 projects, we will only require a little over 29 square miles of infrastructure or a plot about 5.4 miles on a side.

Spreading that out over all 50 states (ignoring the fact that population is not evenly distributed), we would need only ~.6 square miles per state.  Every city of decent size in this country has industrial parks bigger than that.  No problem there.

Taking the this approximation to the extreme -- say, to the "personal fermenter and distiller" high school science project -- dividing the 29 square milles by the 2008 US population of about 300,000,000 gives a silly figure of 10^-7 square miles per person; that's about a foot and a half on a side.  Switching to more rational units, it is ~40 cm on a side.  A family of four (on average) would therefore require roughly a square meter to produce a daily supply of fuel at present consumption levels.  Coincidentally, photos of the efuel100 Microfueler suggest it has a footprint of about a meter square.

Of course, only about two-thirds of total oil consumption goes to transportation, with much of that used by commercial operations, so that family of four would be overproducing even at a meter square (in the present ridiculous units of [production/day/person/area]).  Realistically, larger facilities would probably be employed to produce fuel or "renewable petroleum" for industrial purpposes.

How much the capital costs would be for the square meter of production capacity is up in the air.  The Microfueler lists at ~$10K.  I'll bet the high school students can beat that.

What's yours is yours...right?

Does information describing the pattern of genetic markers embedded in your genome, and even the sequence of your own DNA, belong to you?  I would say yes, but evidently the California Department of Public Health (DPH) has its doubts.

As reported in the LA Times, the DPH has sent cease and desist letters to 13 companies that offer direct-to-consumer genetic testing.  I am especially confused about this because if you have an extra $1-10 million sitting around, you can FedEx your DNA to any number of sequencing companies and have them send you an electronic copy of your sequence in a few months (see my earlier post, "The Million Dollar Genome").

The LA Times, The San Jose Mercury News, and The San Francisco Chronicle all report that the letters were sent following "consumer complaints" about "the price and accuracy of the results".  According to the Chronicle, "California law requirescompanies that conduct genetic testing to have those tests ordered by a licensed physician and to use laboratories that are both licensed by the state and have federal certification."

There appears to be some tension between the interpretation of tests ordered for diagnostic purposes, which probably should require a prescription, and sequencing or genotyping services that provide information about a consumer's genetic makeup.

From the Mercury News:

A spokeswoman for 23andMe, which has financial backing from Google Inc. and Genentech Inc., described the company as an "informational service."

"What we do is offer people information about their genetic makeup, including ancestry and applicable scientific research," spokeswoman Rachel Cohen said.

If physical or pharmaceutical intervention of some sort will be based the results of the test, you probably want a doctor involved in interpreting the results, particularly since correlations between genome sequence and health are still being elucidated.  But even when such correlations are strong, practicing physicians may not know what to do with the information.  As the LA Times points out, "Public health officials have urged consumers to be skeptical, pointing out that most of the research is in its earliest stages and that doctors have little training in interpreting the results."

This gets to the heart of the matter for people interested in knowing their own sequence.  It may be true that connections between relating sequence information and physiology may be sparse, but should that prevent consumers from having access to the raw information?  A physician may take some time to integrate genetic testing into daily practice: should we all be forced to wait until doctors are up to speed?  And what if you just want to know about the pattern of mutations that gives you insight into your ancestry, or are simply curious about the sequence of your own DNA?

Over at Wired News, Thomas Goetz has a few things to say on these issues to the California DPH:

[The cease and desist letters reflect] as much a cultural disagreement as a legal or regulatory one. That is, there is the assumption in the states' letters that, because genetic information has medical implications, the dissemination of this information must fall under their jurisdiction.

But there are, in fact, all sorts of areas in life that have medical implications that we don't consider the province of government -- a pregnancy test, most obviously. We neither want nor assume that doctors should have a gatekeeper role in establishing whether we are or are not pregnant, nor do we look to the state to protect us from that information. Pregnancy is a part of life, and it has all sorts of implications and ramifications. So too with DNA.

For Goetz, who reported for Wired last year on direct-to-consumer genetic testing, the DPH is inserting its bureaucratic nose, and a physician, where neither are wanted or needed:

This is not a dark art, province of the select few, as many physicians would have it. This is data. This is who I am. Frankly, it's insulting and a curtailment of my rights to put a gatekeeper between me and my DNA.

This is *my* data, not a doctor's. Please, send in your regulators when a doctor needs to cut me open, or even draw my blood. Regulation should protect me from bodily harm and injury, not from information that's mine to begin with.

So, bringing this back to the motivations for the cease and desist letters, what of the complaints about "price" and "accuracy"?

The 23andMe homepage advertises that the company provides:

A web-based service that helps you read and understand your DNA. After providing a saliva sample using an at-home kit, you can use our interactive tools to shed new light on your distant ancestors, your close family and most of all, yourself.

Nothing about diagnostics there.  But following the "Gene Journal" link leads to an "Odds Calculator" that will:

Help you put it all in perspective, allowing you to combine genetic information, age, and ethnicity to get an idea of which common health concerns are most likely to affect a person with your genetic profile. While the Odds Calculator is neither a medical diagnostic nor a substitute for medical advice, it can help you confront the bewildering array of health news reported in the mass media and help you decide where you may want to focus your attention.

(Note the specific caveat that the service is not "a medical diagnostic".)

Given the early stage of most efforts to link genomes with physiology, it would be very surprising if a small start-up could assemble the resources to "put it all in perspective".  But even if they don't have the ability to pull that off in a manner I would be satisfied with, it isn't clear that the state should be denying them the opportunity to try.

With respect to the "price" complaint, the last time I checked we are living in a society in which goods and services are priced according to what the market can bear.  Since neither private insurers nor the government is paying for these particular services, which are not intended to provide information to be used in healthcare, there does not appear to be a good argument that the state should care what the price is.

With respect to the "accuracy" complaint, it would seem that these companies are already trying to do business in a competitive environment -- if they aren't providing accurate information then presumably they will succumb to companies that provide better information to consumers.  Again, since this isn't a diagnostic service, it is not clear that the state should intrude in the transaction.

There are a great many snake oil peddlers and quacks out there who offer no caveats as to accuracy or effectiveness, and in comparison 23andMe and its competitors appear paragons of virtue.  Direct-to-consumer genetic information services are creating a new market, and there always bumps along the way in that endeavor, particularly when regulators decide they know more about technology than do innovators.  But it is a market. It is not, in priciple, directly related to health.  Caveat emptor.  Since when is this the concern of Department of Public Health?

Biodesic: It's Alive!

This little post serves as the official launch of Biodesic.  As the book is finally done, or at least mostly in the hands of the publisher, I can turn my full attention to getting the start-up company out of my garage.

As the website says:

Biodesic is part of the new bio-economy. We provide technologies and knowledge to organizations building the future.

Our mission is to transform business and society through the development and distribution of biological technologies.

Here are examples of recent consulting projects.   Our first product is a parallel protein detection tool.  It is similar to Tadpoles but detection is much simpler, and there is no need to amplify a signal using PCR.  We believe the technology will provide novel and useful capabilities for clinical diagnostics and for engineering biological systems.  For more details, see "Technology for Sensitive Multiplexed Protein Detection".