Skip to content

More on the real cost of food

This is another response to the ongoing thread over here. The author (MMM) revised his response pretty heavily while I was writing this, so what I wrote is now less a direct response to his writing, and more of an essay of its’ own. For correct context, here is MMM’s original comment, as I replied to it:

I actually agree more than you might think with the local food movement, although for different reasons: I am more excited about people knowing the people they do business with, because it tends to spark unusual benefits in the happiness of a community.

As more of an economics and capitalism-minded person, I still believe that the highest efficiency is attained with international trade, which overall leads to the highest level of material wealth. If continued material expansion were our goal, and we had the political will to fix the current environmental and labor exploitation side effects, this would be the “richest” situation even after we made it 100% sustainable.

BUT – we passed the level of having enough wealth long ago. So we can afford to make some trade-offs for happiness, which means hanging around and telling jokes with people who live near you. This leads me directly into your camp of going to a simpler and more local lifestyle, including food.

But I have a challenge for you. To make your argument and even convince your capitalism-and-trade-minded adversaries, you need to understand things from their perspective. That means you must understand why “market price” is indeed a good estimate of the social cost of something as long as the environmental and labor rules are fair. And since they are not currently fair, you need to use your imagination to see the good that is mixed in with the bad.

For example: my old high-tech company moved some of its manufacturing to Mexico because their labor costs were 75% lower than what they were paying in Texas. Is this exploitation? I had the chance to go and visit the actual factory and work with my compadres for a week in the new manufacturing plant. They were universally excited about their new jobs, they were earning more than they ever had before in Guadalajara, and using the money to get their kids off to a good start in life, live in a safe place, eat healthier food, etc. These manufacturing plants were a transforming force for Guadalajara and they fed new skills into the city and led to Mexican engineers starting their own companies nearby both partner with and compete with the US-based firms.

That’s is what international trade feels like to MOST of its participants. As I said, your assignment is to read more about it and learn why the argument about “But American jobs are then LOST in the long run!!” is false.

You bring up a few points that I would like to address separately. Broadly speaking though, my belief in a local food system has nothing to do with a desire for protectionism for American jobs. Either I accidentally implied something I did not mean to, or you are incorrectly attributing the arguments that others have to me. I fully understand the concept that “losing” manufacturing and other middle- to low- skilled jobs to markets overseas frees resources here to allow people to move into more highly-skilled occupations. I also believe that system sees diminishing returns at the extreme end of it and we should not go there, but we aren’t there yet. There’s also a somewhat cynical argument that allowing one person to fall out of the middle class here allows half a dozen to be lifted out of poverty overseas, but let’s not go there for now.

The first thing I want to discuss in depth is more about the “price as a measure of social cost” concept. I have done a bit of study on economics, and I’ve never seen this assertion made outside of the article that the previous commenter pointed out. I think there is one or two things going on here that may have led to that assertion.

One might be a problem of perspective. A lot of the economics writing I’ve read talks about the “price” of bringing things to market. In those cases, they seem to me to be talking about the price to the producer, not the price to the consumer. In that context, what they call “price” most people would call “cost”, and the “price” to the consumer is some percentage above the “cost”. By using “price” and “cost” nearly interchangeably without explicit perspectives, the concept of the “cost of production” (which many models equate to the social cost of the object) and that of the “price the consumer pays” are being confused.

The second possibility is that of a logical fallacy. If we accept no confusion between the terms “cost” and “price”, and we accept that price is a function of the marginal cost of an item, the seemingly logical conclusion is that the social cost (which is a function of the marginal cost) would also be closely tied to the price. This is not a safe assumption. At this point we introduce the externalities of pollution and resource depletion, subsidies, and labor force exploitation which makes that relationship even more loose.

To speak in terms meaningful to the vast majority of people, and the terms the apparent originator of the “price = social cost” argument intended we have to look at the price that the consumer pays for an item. That price has at best a tenuous relation to the cost of production. You only have to look so far as loss leaders at the grocery store to see concrete examples of this. It also exists in the consumer electronics world. It has been estimated that Sony was spending $800 to produce Playstation 3 consoles at launch, yet they were being sold at $500. More recently, the Kindle Fire is estimated to cost Amazon about $210 to produce, and yet it is being sold at $199. Both of these are cases where the producer has decided to sell the eco-system forming product at a loss on the expectation that the loss will be recouped in the form of future media sales. Even without attempting to consider externalities, this illustrates concretely that the price of an object is not necessarily even a good estimator of its’ marginal cost, much less the social cost.

In fact, in my research on this subject, I’ve only been able to find two pieces of work that attempt to address the issue of social cost in economics, and both of them come to the conclusion that our current economic models do not adequately account for externalities and are wholly inadequate for measuring social cost! In his conclusion to “Shadow Pricing in Economics”, David A. Starrett ( states, “At this point, we are very far from the capacity to construct such a model, much less analyze it. Many elements in the system interaction are unknown, and the state of the art in computation limits the number of stocks to a handful at best. The challenge now is to construct meaningful models that capture important components of interaction. I hope and believe that ecologists’ experience in modeling the natural world can be a valuable input to this endeavor.”. In “Pollution Externalities: Social Cost and Strict Liability” Peter Lewin ( ) opens with, “This paper critically considers the neoclassical social-cost approach to problems of pollution. This traditional approach, when subjected to close scrutiny, is found to be seriously wanting in applicability and consistency.”. He then goes on to propose a model of “strict liability” to account for this.

Even this basic microeconomics site ( makes no attempt to tie price to marginal cost, much less social cost, beyond the assertion that if price equals marginal cost, total economic welfare is maximized. They do, however, acknowledge towards the end that social cost and marginal cost are different things, and that “A private producer who opts to ignore the negative production externalities might choose to maximise their own profits…”.

All of this confirms my belief that consumer price has no direct bearing on social cost, we lack the models to even conceptualize what that relationship could look like because the systems are too complex, and that it’s perfectly realistic for a producer to ignore externalities they don’t care about.

Next, you provide as a counter-example your experience with high-tech manufacturing in Mexico. Although there are exceptions (the recent suicides at Foxconn’s plants in China are an example of this) I would grant that high tech manufacturing endeavors are generally economically and socially fair to all the participants. Generally, people who will be interacting with that system are well educated and economically independent enough to avoid exploitation. Again, the Chinese model of self-contained manufacturing “cities” where people work and live within the confines of the company complex are likely exceptions to this rule. But that aside, I will grant that in this case everything I’ve learned about this system leads me to believe that we are doing it fairly well. The cost transference (via externalities) I talk about is minimized in the common case. Because your example only deals with the externality of labor cost, I will constrain my point to that area as well.

As you descend from the heights of high-tech manufacturing, down to less skilled occupations, the picture changes substantially. The people involved are less educated, less mobile, and generally more vulnerable to exploitation. The east-asian sweatshop producing clothing is the classical example. Once you get to food production (ie -harvesting) you have even more examples to choose from:

Tomatoes (in America, no less) –
Chocolate –
Bananas –

In the absence of forces to prevent it, the trend for large scale capitalistic enterprises is towards exploitation. Food production is an enterprise that is particularly amenable to this exploitation thanks to the low educational requirements of the participants, and the generally regulation-free (or even despotic) environments of third world countries which are ideal locales for year-round corporate agriculture. This represents a significant external cost to centralized corporately produced food that is not reflected in the price. Even ignoring the environmental costs and other social costs (notably government subsidies), that cost is sufficient in my mind to avoid food produced in that manner whenever possible.

Finally, I’d like to pose a question to you on efficiency. It’s generally accepted that efficiency is good, and maximizing efficiency also maximizes wealth. This relationship could even be said to drive most economic decisions and models at a high level. However, what other attribute do highly efficient systems have in common? By and large, they are fragile. By maximizing the efficiency of food production, what fragility have we introduced? Can we afford to have a system that is that fragile? One where a single producer can accidentally (or heaven forbid, willfully) contaminate the food supply for hundreds of thousands, or even millions, of people? You don’t have to look far for examples of this happening. How can we even begin to account for this in an economic model? At what point does the risk posed by this fragility trump the benefits made possible by ever improving efficiency?

Posted in Life.