Regular reader Kevin Corcoran sent me another email that’s worth posting as a standalone blog post. Here it is:
Earlier this morning I decided to re-read Ludwig von Mises’s book Theory and History: An Interpretation of Social and Economic Evolution. (I know it’s considered bad manners to flaunt my crazy, party-hard lifestyle like this, but hey, it’s the weekend and I like to indulge.) I first read it nearly 20 years ago, and thought it was worth revisiting as I suspect I’ll be able to understand and engage with it much better this time around.
Mises begins by defending the idea of methodological dualism – the idea that the study of social sciences like economics requires a fundamentally different approach than is used in the physical sciences. He points out that physical sciences give way to regularity and prediction that simply isn’t available in social sciences. He writes “Under identical conditions stones always react to the same stimuli in the same way; we can learn something about these regular patterns of reacting, and we can make use of this knowledge in directing our actions towards definite goals…A stone is a thing that reacts in a definite way.” However, Mises argues, there is a crucial difference between the things studied by the physical sciences and the subject of social sciences – humanity. “Men react to the same stimuli in different ways, and the same man at different instants of time may react in ways different from his previous or later conduct. It is impossible to group men into classes whose members always react in the same way.”
Mises doesn’t think this means social sciences are incapable of making predictions of any sort – only that these predictions are only vaguely definable and partly knowable. In his words, “This is not to say that future human actions are totally unpredictable. They can, in a certain way, be anticipated to some extent.” But this partial understanding and vague predictability makes attempts at applied social science fundamentally different from other applied sciences.
One lesson I think we can take from this as economists is to temper the specificity of our predictions to policy changes. For example, when asked what to expect from a binding wage floor, many economists (I count myself among the guilty) have a knee-jerk reaction to just say “increased unemployment.” [DRH note: I always say “reduce employment” because of how unemployment is measured. To be unemployed in the official statistics, you must be out of work and looking for work. The minimum wage law may discouraged those whom it puts out of work, to the point where they don’t bother looking.] But we also realize that this is just one of many ways employers can react. Employers might cut jobs or hold back on creating new positions they otherwise would have. But they often can, and do, respond in other ways. They might cut hours. They might reduce benefits. They might put less effort into providing a pleasant working environment. I thought back to 2014 when minimum wages were being pushed up in the SeaTac area in Washington, and the experiences being described by some workers who “benefitted” from that wage increase:
“Are you happy with the $15 wage?” I asked the full-time cleaning lady.
“It sounds good, but it’s not good,” the woman said.
“Why?” I asked.
“I lost my 401k, health insurance, paid holiday, and vacation,” she responded. “No more free food,” she added.
The hotel used to feed her. Now, she has to bring her own food. Also, no overtime, she said. She used to work extra hours and received overtime pay.
What else? I asked.
“I have to pay for parking,” she said.
I then asked the part-time waitress, who was part of the catering staff.
“Yes, I’ve got $15 an hour, but all my tips are now much less,” she said. Before the new wage law was implemented, her hourly wage was $7. But her tips added to more than $15 an hour. Yes, she used to receive free food and parking. Now, she has to bring her own food and pay for parking.These kinds of experiences are overlooked when economists are too quick to predict unemployment. And when unemployment is given as the predicted cost of binding wage floors, rather than described as one of a variety of different ways employers might respond, situations like the one described can be overlooked. Skeptics might look around, see that technically no jobs were lost, and conclude that economists are wrong to warn about the costs and tradeoffs of this kind of wage setting. It would be more accurate, and more honest, for economists to say “There are a lot of different ways the labor market might adjust. Here are several different ones. I don’t know which employers will react in which ways, and neither do the advocates of this policy. But these increased costs are going to have to be borne by someone, and though the legal incidence of the cost increase is on the employer, much of the economic cost will be borne by the employees, whatever form that cost may take.” [DRH note: As I have often put it, employers adjust on many margins.] Granted, this doesn’t make a quippy sound bite for talk radio, but it has the virtue of being true.
Here is the biography of von Mises in David R. Henderson, ed., The Concise Encyclopedia of Economics and the pic above is of von Mises.
READER COMMENTS
Jon Murphy
Jul 30 2022 at 8:31am
Good stuff from Kevin. His is a point I always stress in my classes: we have to look at the margins people adjust along. Your typical supply and demand diagram assumes the margin is quantity. But, in reality, there are many margins individuals adjust along. Ultimately, this is a major reason I am skeptical of all sorts of interventions, including carbon taxes.
Greg G
Jul 31 2022 at 7:21am
Jon, I take your point (and the point of David’s post) that policy interventions often result in unanticipated and unwanted consequences. All useful prediction in the social sciences is risky. And yet we still have to make decisions about public policy. A decision not to do something still carries an implicit prediction that inaction will lead to a better result than whatever the proposed policy change is. It’s not like only one side in a policy debate is making a prediction.
Can you say more about possible unanticipated adjustments to carbon taxes that make you skeptical of them? I am sure you’re aware that supporters of carbon taxes like them precisely because they allow for more of a role for the market in choosing what the alternatives to fossil fuels will be than many other carbon reduction strategies do.
Kevin Corcoran
Jul 31 2022 at 8:25am
Hey Greg –
Obviously I’m not answering for Jon here – I’m sure he has his own answers to your questions. But I do want to express a point of disagreement, when you say “A decision not to do something still carries an implicit prediction that inaction will lead to a better result than whatever the proposed policy change is.” I disagree, for a few reasons.
First, for what it’s worth, I do think there are in fact good reasons why there should be a strong presumption that inaction usually does lead to better results in many (most?) cases. Michael Huemer said a lot of this well in his paper In Praise of Passivity, where he starts by talking about doctors failed attempts to cure George Washington of illness, and how these attempts were very likely what killed him:
He argues that this is similarly true of societies and civilizations – they, too, are extremely complex, and the scale of how different social forces operate and interact vastly exceeds anyone’s knowledge or understanding:
I agree with Huemer that by default, there are more ways to make things worse than to make things better, both in bodily as well as social health. And Huemer also cites a wealth of evidence about how experts advocating for intervention tend to be wildly overoptimistic and overconfident about the prospect that their proposed intervention will actually have the desired effect. For example, in cases where experts predicted an outcome with 100% certainty, they still turned out to be wrong 80% of the time. This, too, should shift priors towards a great deal of skepticism about the claimed benefits of proposed interventions.
But still, the disagreement I have goes a step further. One doesn’t need to positively assert that nonintervention will produce better results than a specific intervention in order to favor the former. I think the burden is distributed differently – it’s intervention which requires positive justification, and it needs to meet a very high bar, particularly when one considers how prone advocates are to overconfidence. Huemer puts it this way:
I think this argument succeeds in explaining why you don’t need to positively insist “nonintervention produces better results in this given instance” in order to have a default position against intervention. The burden lies with the interventionist, not the other way around.
Greg G
Jul 31 2022 at 9:25am
Thank you Kevin. You make an excellent point. There are many more interventions that can screw up complex systems than can make them better. This is why most, but not all, genetic mutations are harmful. This is why Tolstoy famously said: “Happy families are all alike but each unhappy family is in happy in its own way.”
The issue of a carbon tax is not quite so simple though. We have good evidence that inaction will not just leave things as they are but actually lead to a worsening situation. We are seeing unintended and unwanted consequences of inaction on carbon a regular basis. The idea of a carbon tax is that it makes all non-carbon energy strategies more attractive without imposing one particular solution. The market then can sort out the most efficient response without having one imposed top down.
Free market economists are always the first to tell us that if you tax something you will get less of it. Well, yes! So then don’t be surprised if those listening consider taxing it when they want less of something.
Justice Holmes famously said, “General principles do not decide concrete cases.” There are always multiple competing general principles at play in the interesting cases.
Kevin Corcoran
Jul 31 2022 at 2:42pm
Hey Greg –
Thanks for the reply. I do want to clear up that my point wasn’t about the merits or lack thereof about carbon taxes per se. My main point of pushback was on the idea that lack of intervention needs to be actively justified on the grounds that it would produce better results that some form of active intervention. It’s intervention, not the lack thereof, which actively requires justification on the grounds that it’s highly likely to make a substantial improvement. One can certainly make the case that carbon taxes succeed in clearing that bar – my only point was that the need for justification between intervention and nonintervention is not symmetrical in the way you seemed to suggest.
I have always liked that Tolstoy quote. People can and do interpret it in different ways, but to me it’s always meant something like the following. There are innumerable different ways to live a life that ends up feeling hollow, empty, or unfulfilling. There are very few ways to live which turn out to be compatible with a satisfying and meaningful human existence. And since human nature if very real and not particularly mutable, the ways to live that have turned out to be most fulfilling aren’t likely to change or be superseded anytime soon,
Regarding your quote from Holmes, I prefer the approach taken by Hayek. In the first volume of Law, Legislation, and Liberty, Hayek argues that we should in fact use general principles to decide concrete cases. When people want to depart from a general principle, it’s because they are anticipating some specific and desirable benefit will result. By contrast, holding to the general principle would mean forgoing the specific, anticipated benefit without being able to specify the benefits derived from holding to the principle in this specific instance. As a result, at every point, the desire will be to break away from the principle in each concrete case, and in the end the principle itself is destroyed. As Hayek wrote:
(One day I’ll learn to write short comments, but today is not that day!)
Greg G
Jul 31 2022 at 3:29pm
I’m sure we agree on much more than we disagree on Kevin including most of your reply and I thought the quality of your reply easily justified its length.
I don’t think Holmes was dismissing the importance of general principles. I think he was being honest in pointing out that difficult cases always involve a conflict between important general principles and therefore judging them is more complicated than simply invoking the general principle that leads to the result that you want.
Here is another relevant Holmes quote (and also one from his dissent in Lochner):
“…a constitution is not intended to embody a particular economic theory, whether of paternalism and the organic relation of the citizen to the State or of laissez faire. It is made for people of fundamentally differing views…”
These fundamentally differing views include different concepts of freedom and different views of which general principles are most relevant to specific cases.
Jon Murphy
Jul 31 2022 at 9:01am
I generally support Kevin’s comment above (I’m a big proponent of giving the status quo a strong bias and that reforms of any kind [including liberty oriented] must clear a high bar. Mere speculation that X is better than Y is insufficient to recommend X over Y). So, let me focus on your main question:
Compared to other government-imposed reduction strategies, carbon taxes may be superior. But the argument for them, even in 2022, generally ignores the literature on externalities over the past 100 years. The argument acts like Pigou is the authoritative statement and folks like Coase, Buchanan, Dahlman, Alchian, Demsetz, etc., are generally irrelevant.*
Carbon taxes follow the same logic as standard supply & demand analysis: there are just two relevant variables of price and quantity. Raise the price and you lower the quantity (all else held equal). But, in reality, there are many ways one can react to a price increase that is not reducing quantity. Indeed, it is at lease theoretically possible (though I don’t think likely) that a carbon tax could raise carbon output if carbon for the firm follows the pattern of a Giffen Good. It’ll all depend on what are the relevant alternatives.
*I heard a talk once where the presenter was discussing externalities. One person in the audience asked about Coase’s insights. The presenter just said “Coase only works in a zero-transaction cost world. But this is the real world.”
Knut P. Heen
Aug 2 2022 at 5:08am
There are many problems with carbon taxes. First, carbon taxes will move production to countries that don’t impose carbon taxes thus increasing emissions through transportation. Second, it is generally very difficult to find the carbon content of goods produced, even more so if produced abroad. How much of the carbon emissions from heating the barn and feeding the pigs went into the ham and the bacon? Joint costs cannot be assigned to specific products. A large fraction of energy used in production and transportation are joint costs. These will be assigned rather arbitrarily to each product by cost accountants. Third, carbon taxes becomes someone’s income. This group will lobby for maximizing the revenue of the tax rather than having a tax which reduce emissions optimally. Do carbon taxes work on the public sector? They pay to themselves. Why not set a tax which maximize revenue from the private sector?
A carbon tax needs to be world-wide, calculated correctly, and the revenue from the tax must be burned such that it feels like a real cost rather than a fictional cost to everyone. That will never happen.
Thomas Lee Hutcheson
Jul 31 2022 at 11:17am
However adjustment to a minimum, it’s probably worse than the same income transfer done with higher EITC.
Stéphane Couvreur
Jul 31 2022 at 12:51pm
Thanks for reminding us of this great and often underestimated book by Mises!
The same point is also well made by Gordon Tullock in an article entitled “The minimum wage: A new perspective on an old policy” published in The best of the new world of economics-and then some 5th ed (1989).
I was surprised to learn that Tullock had written about the minimum wage-the usual reference is Stigler.
I was even more surprised by the introduction of his article where he claims-roughly-that “everybody knows that a moderate minimum wage has little or no effect on the volume of employment”. Note that this was written before Card & Krueger.
He then goes on to argue that the worker’s real compensation is comprised of things other than the monetary wage, so that even the workers who keep their job under a minimum wage lose in terms of real compensation because of the wage floor, through reduced benefits, having to put in more effort, etc.
A short version could be “the quality of the jobs decreases even when
-by chance-their quantity doesn’t”. It is easy enough to understand how a price ceiling on bread can decrease the quality of the baguette; here the same reasoning applies to a wage floor.
Comments are closed.