Does government intervention create its own externalities (or “neighborhood effects”)? Many economists think so and the question appears especially important in the current storm of dirigisme. Sixty years ago, Milton Friedman defended the idea in his influential book Capitalism and Freedom. A mainstream neoclassical economist and moderate classical liberal, Friedman wrote (p. 32):
Our principles offer no hard and fast line how far it is appropriate to use government to accomplish jointly what is difficult or impossible for us to accomplish separately through strictly voluntary exchange. In any particular case of proposed intervention, we must make up a balance sheet, listing separately the advantages and disadvantages. Our principles tell us what items to put on one side and what items on the other and they give us some basis for attaching importance to the different items. In particular, we shall always want to enter on the liability side of any proposed government intervention, its neighborhood effects in threatening freedom, and give this effect considerable weight. Just how much weight to give to it, as to other items, depends upon the circumstances. If, for example, existing government intervention is minor, we shall attach a smaller weight to the negative effects of additional government intervention.
This is an important reason why many earlier liberals, like Henry Simons, writing at t time when government was small by today’s standards, were willing to have government undertake activities that today’s liberals would not accept now that government has become so overgrown.
Technically, externalities are usually modeled as non-intentional effects of activities carried on for other purposes. Otherwise, everything that imposes indirect costs or benefits on somebody would be an externality; pretty much all activities would fall in that category. It seems to follow that the typical government intervention should not count as a positive or negative externality, because it is explicitly designed to create benefits for some groups and impose corresponding costs on others. However, if it also has indirect consequences on everybody’s liberty, it can be considered as creating freedom externalities, as Friedman suggests. (In this perspective, a government intervention whose purpose is to increase government power and to decrease individual liberty would not generate freedom externalities, but only direct freedom costs.)
Does growing government intervention, besides increasing freedom externalities, also increase their rate of increase, as the Friedman criterion above seems to say? For any individual, the cost of a given intervention in terms of his own individual liberty will conceivably be larger the higher is the starting level of government intervention and power. One reason would be that, at higher thresholds of power, the more likely an additional intervention will combine with existing controls to give irresistible power to government and seriously undermine the liberty of the subject (or “citizen”). If government surveillance is widespread, for example, the more likely a new public morality or lifestyle law can be used to harass unpopular minorities. Another reason is simply that, as individual liberty decreases, the more an individual will find the remainder valuable.
Note how in order to avoid the serious problem of cost-benefit analysis—which is that no scientific basis exists for weighing the benefits of some individuals against the costs imposed on others—we should formulate the problem of freedom externalities à la James Buchanan: each individual estimates his own cost and benefit from a given intervention and can be presumed to consent to it only if, for him, the latter is larger than the former. The only assumption made here is that, everything else equal, no individual wants to be more oppressed; oppression is a cost, not a beneficial or neutral condition. If some individuals like to be slaves for the mere pleasure of servitude, freedom externalities are not unambiguously positive or negative. The problem then becomes more complicated.
Considering only negative freedom externalities, Friedman’s warning is valid: the higher the level of government intervention, the larger are the negative freedom externalities of any new proposed intervention. I suggest that it is not easy today to find any new government intervention—or at least any “net” intervention—that would survive the Friedman criterion.
READER COMMENTS
Thomas Lee Hutcheson
Jul 10 2022 at 4:29pm
“Technically, externalities are usually modelled as non-intentional effects of activities carried on for other purposes. Otherwise, everything that imposes indirect costs or benefits on somebody would be an externality; pretty much all activities would fall in that category.”
I’m not sure “intention” is the key determinant. I think is more a matter of having an un-market mediated effect. Whether I “intend” to raise the CO2concentration of the atmosphere when I turn on my gas oven, the effect on the sea level 40 years from now is the same.
As for Friedman, I think that “freedom” is largely accounted for by counting economic costs of an intervention. It being so hard to get real economic cost benefit-analysis of regulation done, I doubt trying to add “freedom” would help much, especially since in practice most regulation is consideration of one kind of market intervention rather than another. Wouldn’t a tax on net CO2 emissions infringe less on freedom that a set of subsidies to zero and negative CO2 technologies financed who knows how that added up to the same reduction in CO2 concentration? And I am not at all troubled by saying that adding and subtracting dollars of cost and benefits is “unscientific.”
Pierre Lemieux
Jul 10 2022 at 10:26pm
Thomas: My claim is that intentionality is one necessary condition for a useful concept. See my answer to Mark below.
On the question in your penultimate sentence, I mentioned “‘net’ intervention” precisely to answer it in advance (even if it makes the Friedman argument and mine less neat). In practice, it is likely that, with a non-angelic government, a carbon tax would add to subsidies and regulations: just look at Europe.
Thomas Lee Hutcheson
Jul 11 2022 at 4:51pm
I’d ask you to look more broadly, even leaving aside that Europe does not have a tax on net CO2 emissions.
I am thinking about the future. Many people believe that we (sorry about the “we” :)) need to adopt policies to slow the increase and eventually reduce the concentration of CO2 in the atmosphere. Many have already been adopted. Accepting that a new tax on net CO2 emissions would not wipe the slate of prior policies clean, it still seems reasonable to me to think that it would be to some extent a substitute for more costly policies.
You will probably note the similarity of my argument about the Child Tax Credit/child care credit/child care provision or harking back to an increase in the EITC/increase in the minimum wage.
In general I am trying to help you argue more persuasively with Neoliberals. 🙂
Mark Brady
Jul 10 2022 at 7:32pm
“Technically, externalities are usually modelled as non-intentional effects of activities carried on for other purposes.” No, I’ve never come across such a definition. Some externalities may be non-intentional, but others are intentional. In Modern Principles 5th ed. (2021) Tyler Cowen and Alex Tabarrok define externalities as “external costs or external benefits, that is, costs or benefits, respectively, that fall on bystanders.’ The intention of the economic actors does not enter the definition.
Pierre Lemieux
Jul 10 2022 at 10:15pm
Mark: If you don’t agree, my “usually” was perhaps too strong. Or perhaps the incidental aspect of externalities is known only to experts who have to think hard about the topic? I am interested in anything you know that I don’t, but here is an excerpt of my Regulation article on externalities:
Think of the implications of the contrary view.
Dan Merilatt
Jul 11 2022 at 8:16am
If, as I understand Friedman and you, incremental government expansion is increasingly deleterious to individual freedom the larger the government sector; would it therefore not be easier to persuade people to support reductions in the size and scope of government? Alas that, in my experience, does not seem to be the case.
Pierre Lemieux
Jul 11 2022 at 3:11pm
Good question, Dan! Part of the answer is that the individual voter, who has no chance of influencing the outcome of an election (contrary to the outcome of his choice if he decides to by a SUV), remains rationally ignorant of politics and, in two thirds of the cases, rationally apathetic. Another part of the answer is government propaganda. And remember Joseph Schumpeter in Capitalism, Socialism and Democracy:
Craig
Jul 11 2022 at 9:24am
“Does growing government intervention, besides increasing freedom externalities, also increase their rate of increase, as the Friedman criterion above seems to say?”
An excellent example of this is the Vietnam War. Ignoring the direct costs of the war for one moment, the US government ordered Agent Orange which was a defoliant. Makes sense in a jungle environment, but there were far reaching effects from the government ordering this chemical from producers along the Passaic River in Newark, NJ.
Combining that with other EPA/NJDEP regulations, years later the combo created a situation where Newark had a ‘negative land value’ problem. Also they had other concerns about dredging Newark Bay out of a fear that the dredging would dredge up Agent Orange.
Some might say that Newark sat at the edge of the Rust Belt and in a sense that is correct, Newark had been experiencing population declines before the Vietnam War and then post riots experienced further population losses. It wasn’t until the 2020 Census that the population in Newark finally clicked up.
Jose Pablo
Jul 11 2022 at 1:30pm
“Technically, externalities are usually modelled as non-intentional effects of activities carried on for other purposes. (…). It seems to follow that the typical government intervention should not count as a positive or negative externality, because it is explicitly designed to create benefits for some groups and impose corresponding costs on others”
I don’t agree, since most of the cost of government interventions are “unintended”, and so could be used as textbook examples of your definition for “externalities”. I don’t really think Mao was trying to engineer a famine that would kill 30-45 millions of Chinese or that Castro’s intent was forcing millions of Cubans out of Cuba or that the Allende’s government in Chile intention was to create the highest inflation the country had ever seen, or that the intention of the Harold’s government in the UK was to force the collapse of Britain’s industries (I can’t find any example involving the Venezuelan government … sorry)
I would be more precise to say: “It seems to follow that the typical government intervention is the best provider of textbook examples of this definition of externalities since most of the cosenquences of its policies are unintended”
Pierre Lemieux
Jul 11 2022 at 3:26pm
Jose: “Unintended” effects that happen again and again in similar government interventions are not unintended but collateral damages, which are better conceived as intended. If I notice that X% of the times street robbers try to rob somebody, they need to shoot him through the chest, I can’t say that the deaths of robbery victims are unintended effects. Similarly, the loss of jobs caused by the minimum wage (say) are not an externality.
Jose Pablo
Jul 11 2022 at 4:41pm
Fair enough
Jose Pablo
Jul 11 2022 at 1:37pm
“If some individuals like to be slaves for the mere pleasure of servitude, freedom externalities are not unambiguously positive or negative”
It seems to be the case that some individuals do prefer the peace of mind of having others taking care of their needs at the cost of renouncing to their freedom. Why this preference of theirs entitled them to force me to provide this care (something I don’t want to do, particularly if forced to), is a mystery to me (not a fan of Rawls).
Pierre Lemieux
Jul 11 2022 at 3:36pm
Jose: You write:
I wrote “everything else equal” and “for the mere pleasure of servitude” precisely in order to deflect that objection. Everybody makes trade-offs between goods, and I assumed (following the Enlightenment) that freedom is a good for everybody. Whether this assumption is true or not has momentous consequences, as I tried to argue before.
Jose Pablo
Jul 11 2022 at 1:47pm
“cost-benefit analysis —which is that no scientific basis exists for weighing the benefits of some individuals against the costs imposed on other”
Actually, not sound scientific basis exist even for adding intertemporal preferences for just one individual (even if the calculations are done by this very same individual)
The “technic” we usually use is full of theoretical inconsistencies making it closer to a profession of faith than to a scientific exercise. It is, at the end of the day, a “scientific tautology” since people “scientifically” do it just by looking at how other people do it.
The advantage of having the individual doing its own weighting and time discounting is that, in this case, the accountability is pretty straightforward (not a minor one).
Jens
Jul 11 2022 at 3:30pm
If intertemporal preferences for an individual are already difficult for the individual itself to calculate on a scientifically sound basis, does this not possibly also mean that the individual – if understood as self-consciousness reflecting on itself in time and space – cannot even add up its own preferences ?
Jose Pablo
Jul 11 2022 at 4:52pm
It means Jens, that the individual cannot add up its own intertemporal preferences in any “scientifically proof” way.
As a matter of fact, if you let the individual add up its own intertemporal preferences in, let’s say, 3 different moments in time (same information just different “mood”) he could, very well end up with different conclusions (so, his/her conclusions cannot be “reproduced”).
The scientific (so to speak) way of aggregating intertemporal preferences is constructed based on these “volatile conclusions” … and indeed volatile it is!
Pierre Lemieux
Jul 15 2022 at 7:21am
Jens: In modern economics, preferences are just an ordinal ranking. The only thing the individual needs to know is if he prefers, say, a big car and a small house or a small car and a big house. He does not have to add up his preferences, which indeed cannot be subject to arithmetic operations.
Comments are closed.