Which other government programs are us fat?

A few days ago, I took on the claim that farm subsidies are making us fat (the answer is most likely "no").  However, there are other government programs that potentially affect food prices - what about those programs?  Have they contributed to the rise in obesity?

A new paper in by Julian Alston, Joanna MacEwan, and Abigail Okrent in Applied Economic Perspectives and Policy asks whether funding for agricultural research and development (R&D) can explain the rise in obesity.  The chain of logic goes like this: there is extensive evidence that funding for agricultural research increases productivity; higher productivity means getting more food using fewer resources; more food means lower food prices; more food at lower prices means more food intake; more food intake leads to obesity.  Ergo, government funding for agricultural research leads to obesity.  

So what did the authors find?  They found that agricultural R&D spending probably did have a modest effect on obesity rates, but that R&D also resulted in enormous benefits to consumers and producers.  The authors write:

“Our analysis of historical counterfactuals suggests that it would have been very expensive to have foregone past R&D-induced productivity growth, even if in doing so we were able to reduce obesity and related healthcare expenditures.”

And, if we had undone the R&D efforts that led to the food price changes since the 1980s:

“This would be a costly reversion; it would cost consumers $65.01 billion, of which only $4.72 billion would be offset by savings in public healthcare costs, to reduce average U.S. adult body weight by 4.85 lbs. This translates to a cost of $55.6 per pound after the savings in public healthcare costs are taken into account.”

In summary:

“The implication is that agricultural R&D policy is unlikely to be an effective policy instrument for reducing obesity, both because the effects are small and because it takes a very long time, measured in decades, for changes in research spending to have their main effects on commodity prices. Moreover, as our results and others have shown, the opportunity costs of reducing agricultural research spending in the hope of eventually reducing the social costs of obesity would be very high because agricultural research yields a very large social payoff.”

Having now discussed the effects of farm subsidies and agricultural research, what about programs like the government-sanctioned check-off programs?  That was the topic of a session at the most recent AAEA meetings in Boston.  Parke Wilde from Tufts and Harry Kaiser from Cornell debated the role of check-off programs and their role in affecting public health and nutrition.  I was unfortunately unable to attend the session, but Parke offered a preview of it on his blog.  I hope to see some research on this topic in the near future.  

Are Farm Subsidies Making Us Fat?

In the past couple weeks, there have been a number of popular press articles suggesting that farm subsidies are a big part of the reason Americans eat unhealthy and are overweight. Here's the title from the New York Times: "How the Government Supports Your Junk Food Habit", and Fox News: "Government heavily subsidizes junk food, report suggests", and NPR: "Does Subsidizing Crops We're Told To Eat Less Of Fatten Us Up?". All the hubbub seems to stem from this article by some CDC researchers in JAMA Internal Medicine, which shows people who are more overweight tend to get more of their calories from foods that happen to be subsidized.

But, as we should all know by now, correlation is not causation.  Here's Tracie McMillian in a piece for National Geographic:

“But what the study does not show is the degree to which subsidies—and, in particular, the ones that are currently in place—actually persuade people to eat more of those foods. The researchers, by the way, admit this: “We cannot say [the link between subsidies and consumption] is causal from this study,” says K.M. Vankat Narayan, a lead author.

So while there’s an accepted correlation between low prices and increased purchases, nobody really knows how much farm subsidies matter when it comes to which foods people buy—and eat.”

She's right on the first part and wrong on the second.  There are actual lots of people who know how much farm subsidies contribute to food consumption, and they're called agricultural economists (in fact, McMillian goes on to then cite two prominent food and agricultural economists on the issue: Parke Wilde and David Just).  My view is in line with Wilde's and Just's:

“Indeed, in contemporary America, “the potential impact of the agricultural subsidies on consumption right now is inconsequential,” argues David Just, an agricultural and behavioral economist at Cornell University. Subsidies for farmers are unlikely to have much impact on consumer prices, adds Wilde, because farmers’ share of what we pay at the store is so little. ”

Let me pause right here and say that the question of the causal relationship between farm policy and unhealthy food consumption is an empirical, positive question, not a normative one.  There are a variety of reasons one may think we should or should not have farm subsidies (I generally find myself in the latter camp for reasons I won't go into here), but for the moment let's set the "should" question aside and ask what the evidence actually says on the link between farm subsides and unhealthy eating.  

Here's what I wrote on the issue in a recent Mercatus paper (which came out well before all the JAMA paper and the resulting news stories):

“Despite popular claims to the contrary, research suggests that farm subsidies have likely had little to no effect on obesity rates. First, although such policies may have had some effect on farm commodity prices, these inputs account for only a small share of the overall retail cost of food. For example, in 2013, only 7 percent of the retail price of bread was a result of the farm-gate price of wheat and other agricultural commodities. Even the enormous price swing that took wheat from around $3 per bushel in 2006 to almost $12 per bushel in February 2008 (a 300 percent increase) would be expected to increase the price of bread by only about 14 percent. Second, agricultural policies are mixed, and some policies (such as those for sugar, ethanol promotion, and the Conservation Reserve Program, or CRP) push the prices of agricultural commodities up rather than down. Third, despite the widely varying agricultural policies across countries and over time (see figures 14–16), those policies do not correlate well with differences in food prices and obesity rates across countries or with changes in obesity rates over time.”

In the model I used for the forthcoming paper I wrote on the distributional impacts of crop insurance subsidies, I find that the complete removal of crop insurance subsidies to farmers would only increase the price of cereal and bakery products by 0.09% and increase the price of meat by 0.5%, and would also increase the price of fruits ad vegetables by 0.7%.  So, while these policies may be inefficient, regressive, and promote regulatory over-reach, their effects on food prices are tiny, and depending on which policy we're talking about, could push prices and consumption  up or down.  

For those truly interested, here's a small list of academic papers by economists on the relationship between farm policy and obesity/health (for links to the actual papers, just do a quick googlescholar search).

Alston, Julian M., Daniel A. Sumner, and Stephen A. Vosti, “Farm Subsidies and Obesity in the United States: National Evidence and International Comparisons,” Food Policy 33, no. 6 (2008): 470–79. 

Balagtas, J.V., Krissoff, B., Lei, L. and Rickard, B.J., 2014. How Has US Farm Policy Influenced Fruit and Vegetable Production?. Applied Economic Perspectives and Policy, 36(2), pp.265-286.

Beghin, John C., and Helen H. Jensen. "Farm policies and added sugars in US diets." Food Policy 33, no. 6 (2008): 480-488.

Miller,J. Coreyand Keith H. Coble, “Cheap Food Policy: Fact or Rhetoric?” Food Policy 32, no. 1 (2007): 98–111. 

Okrent, Abigail M.  and Julian M. Alston, “The Effects of Farm Commodity and Retail Food Policies on Obesity and Economic Welfare in the United States,” American Journal of Agricultural Economics 94, no. 3 (2012): 611–46.

Rickard, B.J., Okrent, A.M. and Alston, J.M., 2013. How have agricultural policies influenced caloric consumption in the United States?. Health Economics, 22(3), pp.316-339.

Zilberman, D., Hochman, G., Rajagopal, D., Sexton, S. and Timilsina, G., The impact of biofuels on commodity food prices: Assessment of findings. American Journal of Agricultural Economics, 95, no. 2 (2013) : 275-281.

Farm Subsidies - Magnitudes and Comparisons

Continuing the discussion of the paper I wrote entitled "The Evolving Role of the USDA in
the Food and Agricultural Economy", today I'll discuss some USDA farm support programs (in another post, I'll discuss the academic research on effects of these and other USDA programs). 

In 2012 (the last date of the Census of Ag), the average government payment per farm receiving payments was $9,925. However, a large percentage of farms receive no government payments . In particular, farms that sell less than $50,000 worth of products tend not to receive payments, while the opposite is true for farms with sales greater than $50,000. For the 3.9 percent of farms with sales of $1 million or more, 71.2 percent receive payments averaging $40,559. Whereas the smallest farms receive the smallest average payments in total dollars, they receive the largest payments when expressed relative to value of production. Farms with sales of less than $1,000 that receive payments tend to get 9.36 cents for every dollar of output produced, but farms with sales of more than $1 million that receive payments tend to get only about 2 cents for every dollar of output produced.

Although government payments represent a small fraction of the value of output (i.e., gross revenue), they are certain to represent a much higher fraction of farmers’ net income. In fact, USDA Census of Agriculture data show that in 2012, the average net cash income for each category of farm with sales of less than $24,999 was negative. Those farms operate at a loss; because of this, whatever government payment they receive is infinitely greater than what they
make from farming. The average payment as a percentage of net income (for those receiving payments) is 31 percent, 18 percent, 13 percent, and 7 percent for farms with total sales in the categories $100,000 to $249,999, $250,000 to $499,999, $500,000 to $999,999, and $1 million or more, respectively.

It is interesting to compare all this with SNAP payments.  As indicated, of the farms receiving payments in 2012, the average payment was $9,925. By contrast, USDA data indicate that the average payment per individual receiving SNAP in 2012 was $133 per month, which amounts to $1,596 annually. SNAP payments increase at a decreasing rate with the size of the household. For a four-person household receiving SNAP benefits, the average payment was $440 per month, or $5,280 per year, in 2012. Food assistance programs represent a larger share of the USDA budget than do farm support programs because SNAP recipients far outnumber the recipients of farm program payments, not because each SNAP recipient receives a higher payout than does each recipient of farm supports.

It is also useful to compare US farm support payments with those in other countries.  For this, we can turn to data from a World Bank project led by the Kym Anderson and colleagues. 

The nominal rate of assistance (NRA) is defined as the percentage increase or decrease in gross returns to farmers caused by government policies. A positive number means a country’s
policies are pushing up agricultural prices and returns, and a negative number implies the opposite. The gross rate of assistance (GRA) is the NRA expressed in absolute dollar terms (in the year 2000) instead of in percentage terms. The GRA is the NRA multiplied by the value of agricultural production in a country divided by the number of farmers.

Figure 14 shows the average NRA, and figure 15 shows the average GRA of 53 different countries from 2000 to 2010. The figures contain a selection of developed and developing countries to provide insight into the diversity of agricultural policies around the world. The United States had an average NRA of 11.2 percent and a GRA of $3,576 per farmer over this period. That means that the gross returns of US farmers are 11.2 percent (or $3,576 per farmer) higher than would have been the case were it not for various government policies. Some countries, such as Norway, Iceland, Switzerland, and the Republic of Korea, have NRAs higher than 100 percent. Thus, US agricultural policies push farmer prices and returns higher than would be the case in the absence of such policies, but by an amount far less than is the case in some other countries and far more than in others.

Whereas figure 15 shows a snapshot of the GRA at a point in time, figure 16 shows changes in the GRA per farmer over time in eight selected locations (all in 2000 dollars). The GRA per farmer in the United States increased sharply from the 1970s to the 1980s and has subsequently stayed around $3,000 per farmer per year. The GRA per farmer in Japan has risen over the entire period considered from only $536 per farmer per year in the 1960s to $8,653 per farmer per year from 2000 to 2009. New Zealand dramatically lowered the GRA per farmer from the 1980s to the 1990s. Brazil and China have policies that are relatively neutral with regard to farmer gross returns. Until recently, countries in the European Union had highly distorting policies equivalent to taxes in excess of 100 percent.

In most locations (except eastern Europe and central Asia), agricultural policies have distorted the overall economy less since the 1980s. From 2000 to 2010, the United States had a welfare reduction index (WRI) of 17; the only locations that had less distorting policies were Australia and New Zealand, which had an average index of only 3.8 over this period (figure 17).

The welfare reduction index (WRI)  accounts not only for transfers but also for trade policies that affect the food and agricultural economy. According to Anderson, Rausser, and Swinnen, the WRI is calculated as “the percentage uniform trade tax which, if applied equally to all agricultural tradables, would generate the same reduction in national economic welfare as the actual intrasectoral structure of distortions to domestic prices of these tradable goods.”

The previous graphs aggregate the effects of agricultural and trade policies across all commodities. Figure 18 shows the average NRA for 11 different commodities in the United States from 2000 to 2010. During that period, sugar, cotton, and milk producers benefited most, with NRAs of 75 percent, 56 percent, and 39 percent, respectively. Barley and wheat had relatively low NRAs. Other commodities like beef and pork (not shown in the graph) had NRAs near zero.

To put these figures in perspective, it is useful to compare them with other distortions in the economy. In a remarkable statement, Anderson, Rausser, and Swinnen write,

“In 2004, existing agricultural and trade policies accounted for an estimated 70 percent of the global welfare cost of all merchandise trade distortions, even though the agricultural sector contributes only 6 percent of global trade and 3 percent of global GDP.”

In short, despite the small contribution of agriculture to global GDP, agricultural policies are responsible for the lion’s share of welfare losses that result from trade distorting policies.

In the paper I also talked about the fact that USDA impacts on the economy likely extend beyond those caused by explicit farm-commodity policies. To get a sense of such impacts, I utilized the RegData database.  You can read the paper for more details on that data set, or look at some of the work by Levi Russel who blogs at FarmerHayek.com.

Zhen et al. on Soda Taxes

Yesterday, I was browsing recent back issues of the American Journal of Agricultural Economics looking for papers on consumer demand (somebody has an AAEA presidential address to write).  

I came across two papers by Chen Zhen and co-authors published in 2014 on effects of sugared sweetened beverage taxes (or "soda taxes") that I'd previously read but not blogged on before.  I thought I'd mention them here given the ongoing policy discussions surrounding the issue (Philadelphia politicians are currently considering a soda tax; Oakland has a ballot measure planned on the issue; and there is much debate about the potential effects of the soda tax already passed in Mexico).

In the first paper, Zhen and colleagues show that the way most of these taxes are designed, on a per ounce of soda basis, is not nearly as effective as would be a tax on a per calorie basis.  The authors write:

“For every 3,500 beverage calories reduced, the estimated consumer surplus loss due to a calorie-based tax is $1.40 lower than the loss induced by an ounce-based tax. A 0.04 cent per kcal SSB tax is predicted to reduce beverage energy from ScanTrack supermarkets by 9.3%, compared with 8.6% from a half-cent per ounce tax. Applying this percentage change to beverages obtained from all sources, we calculated that a 0.04 cent per kcal tax on SSBs will reduce total beverage energy by about 5,800 kcal per capita per year.14 Compared with an ounce-based SSB tax that also achieves a 5,800 kcal reduction in beverage energy, the 0.04 cent per kcal SSB tax is estimated to save $2.35 per capita or $736 million for the U.S. population in consumer surplus per year.”

The "lost consumer surplus" means consumers are worse off with either tax.  This is an issue I've raised several times before: there have been few serious attempts to carefully articulate how a soda tax improves consumer welfare given that consumers don't like paying higher prices (e.g., see here or here).

In the second paper, the authors show how a sugar-sweetened beverage tax might have unintended consequences. 

“The preferred demand specification predicts that almost half of the reduction in SSB calories caused by an increase in SSB prices is compensated for by an increase in calories from other foods. We further found potential unintended consequences of an SSB price increase on sodium and fat intake. Because energy intake is just one of many dimensions of nutrition, the results on sodium and fat highlight the complexity of using targeted food and beverage taxes to improve nutrition outcomes.”

They predict that one half-cent per ounce tax on such beverages would reduce body weight by "0.37 and 0.16 kg/person in 1 year and 0.70 and 0.31 kg/person in 10 years for low- and high-income adults, respectively."

They also write:

“The welfare loss for low-income households is about $5 per household per year more than high-income households because low-income households reported higher SSB purchases in Homescan. This difference in welfare loss between low- and high-income households reinforces the regressive nature of an SSB tax”

Support for GMO Labeling a Left-Wing Phenomenon?

Much has been written about whether aversion to biotechnology and GMOs has ideological dimensions rooted in the left.  I've written about this before, as have many others (this paper in Food Policy extends the discussion to a whole host of food regulations beyond biotechnology).  Most of the studies I've seen (including my own data) suggest only small differences in the left and the right in terms of beliefs about safety of eating GMOs.  However, as I previously argued:

“One distinction, which I think is missing, is the greater willingness of those on the left to regulate on economic issues, such as GMOs, than those on the right. Stated differently, there are questions of science: what are the risks of climate change or eating GMOs. And then there are more normative questions: given said risk, what should we do about it? Even if the left and the right agreed on the level of risk, I don’t think we should expect agreement on political action.”

Perhaps the clearest demonstration of this difference in willingness to regulate comes from a new paper by John Bovay and Julian Alston in the Journal of Agricultural and Resource Economics. They look at precinct-level voting data on the Prop 37 mandatory labeling initiative in California in 2012. One of the best predictors of support for Prop 37?  The share of people in the precinct voting for Obama. Here's a telling graph from their paper.   It's an almost perfect positive, linear relationship. 

The authors went on to use these results to predict what would have happened in other states if they'd had an opportunity to vote on Prop 37 (I should note we did something very similar in a paper on for votes on California's Prop 2 related to animal welfare in 2008).  Bovay and Alston found the following:

“Projections using our estimated model imply that a majority of voters in only three of fifty states (Hawaii, Rhode Island, and Vermont) plus the District of Columbia would have passed Proposition 37 had it been on their ballots in 2012”