The Problem of Retirement As a Window Into the Incoherence of Orthodox Economics

-- Posted by Neil H. Buchanan

My most recent Verdict column, published last Thursday, asks if "conservatives really want people never to retire."  Obviously, I am not claiming to have heard any conservatives say that they do not want people to be able to retire.  Instead, I argue that the inevitable logical extension of their arguments must be either that they do not want anyone to be able to retire, or perhaps that they only want people who are wealthy to be able to retire.
That is, I am not asking, "Can conservatives really mean it when they say that they do not want people to be able to retire?" but rather, "Don't conservative arguments naturally imply that people would not be able to retire, even though conservatives have never said so, and even if they would be uncomfortable with that implication?"  My column shows that, applying standard orthodox economic reasoning, conservatives' logic would, indeed, seem to lead to that unexpected conclusion.

In the recent squabble over whether "Obamacare will kill 2.5 million jobs," a prominent conservative economist was horrified when liberals applauded estimates that some people would be "unlocked" from their jobs by the availability of health care coverage through the Affordable Care Act (ACA).  The Congressional Budget Office had published estimates that, by 2023, people might choose to reduce their labor supply in an amount aggregating to 2.5 million fewer worker-years of work.

This reduction in labor supply would be the result of some unspecified (and unknown, because the economic model used to generate the estimate cannot provide the specific sub-totals) combination of people working fewer hours because of a fear of losing health-care subsidies due to income phaseouts, moving among jobs because people no longer need to stick with a bad job merely to hold onto good (or any) health insurance, or retiring early from jobs because they will no longer need to wait until age 65 to qualify for something like Medicare.

It is that final category (early retirement by physically beaten-down people, who will no longer need to continue to harm themselves by hanging onto jobs because of the medical benefits) that provides the point of departure in my column.  Looking at that category in particular, liberals have said, "And that's good!"  That is, we asked why it would not be a good thing for people to exercise their free choice to end their participation in the paid work force.  The ACA makes it possible to do so, by giving people a broader range of choices than they faced prior to the ACA's enactment.  Conservatives said, in essence, that liberals were wrongly celebrating a distortion in the labor market.  That is, these additional people will only be able to retire because of the ACA, and the ACA is a government policy that distorts the labor market, so those people should not be able to retire.

As always, the "should" in the previous sentence raises a red flag, because orthodox economists' obsessive desire to maintain their scientific veneer requires them ritualistically to deny that they have anything to say about normative matters.  They might (or might not) be able to stop themselves from saying that the increased early retirements are bad, but they can at least claim that the change in the labor market will be inefficient, leaving it to the listener to draw the obvious normative implication.  Maybe, these economists will grudgingly allow, the proverbial "philosophers and politicians" will conclude that there are moral (or simply unprincipled) reasons to tolerate the economic inefficiency that the ACA will surely cause, but the orthodox economic view is that the shift in labor supply is surely evidence of the ACA's economic inefficiency.

As I argue in my column, however, that reaction by conservatives exposes their opportunistic use of the pre-ACA laws as the correct standard of comparison in determining whether a policy is efficient or inefficient.  If the reduction in labor supply is presumptively inefficient, then these economists must have silently assumed that the labor supply without the ACA is efficient.  Or, at the very least, they must be assuming that reducing labor supply moves us in the wrong direction, even further away from some other hypothetical efficient equilibrium, from which real-world policies that have been enacted by stupid politicians have already moved us.

At this point, however, we start pulling on the thread of what it means to have a hypothetical standard of comparison that represents the "no government intervention" world so dearly loved by conservatives.  If their argument is that people should not be able to rely on provisions of the ACA to retire -- and note that I am deliberately dropping the pretense that they are not really saying "should" and "should not" in their pronouncements -- then what should people be allowed to rely upon in their retirement?

Surely, they cannot be allowed to rely on Social Security or Medicare (and definitely not Medicaid, which is means-tested).  "Get your government hands off my Medicare!" is still one of the most revealing moments of ignorance in recent political history, but it was not merely one confused retiree who said it.  One of the most celebrated conservative economists once said: "[J}ust wait till you see Medicare, Medicaid, and health care done by the government."

But the government's retirement programs are not the only things that are out of the picture, in terms of constructing a hypothetical standard of comparison to determine whether a reduction in labor supply is efficient (moving toward that hypothetical standard) or inefficient (moving in the opposite direction).  Government-subsidized retirement savings plans like 401(k)'s are out, because those tax code-based subsidies, by the logic of orthodox economics, would presumptively move the economy away from the efficient amount of saving toward a world with too much saving.

But then, any amount of saving (other than storing physical commodities in fallout shelters) is only going to be possible if financial institutions are forced to honor their legal commitments, by a government that defines and enforces laws of contract that are not given by God.  Certainly, there can be no government-provided deposit insurance, protecting depositors against bank runs.  Pushing the logic yet a further step back, the money that would be deposited into those government-protected accounts would have to come from one of two sources,  either (1) the fruits of one's labor, the wages of which have been paid in full because the government has made it clear that it will enforce at least the most minimal labor laws (unpaid labor being theft, after all), or (2) the proceeds of inheritances and gifts, which exist only due to property laws, estate laws, and so on.

I am, of course, now talking about the familiar "baseline question" that I have discussed frequently here on this blog.  This discussion, however, helps to clarify some confusion in academic discussions about the required role of government in making economic activity possible.  Frequently, people will say that this is "the Murphy/Nagel point," referring to an already-classic 2002 book by the NYU philosophers Liam Murphy and Thomas Nagel, The Myth of Ownership.  Indeed, I have made that error in casual conversation myself.

In fact, Murphy and Nagel's innovation was not the observation that economic activity is only possible when a government exists.  Economists of earlier generations (that is, economists who had not yet completely bought into the current orthodoxy, but instead were willing to think more broadly) were fully aware of the absurdity of talking about "no-government worlds," or "states of nature" where economic activity magically took place without government-created and -enforced rules.  Early legal realists were aware of this as well.  Neither Murphy nor Nagel would claim to have been the first to make this point.

Their argument was that the logical absurdity of talking about a no-government world makes it logically incoherent to talk about (much less measure) "before-tax income."  That is, because there is no economy without a government, and there can be no government (even a minimalist one) without tax revenue, it is not meaningful to ask how much money a person would receive as a result of market transactions, if there were no taxes being collected.  Because different tax laws will change economic behavior in different ways, measurements of pre-tax income will differ, depending upon how the government is presumed to be collecting the money necessary to allow economic activity to proceed in the first place.

This is why so much of the discussion in the Murphy/Nagel book is directed against "street libertarianism," with its claims (by people like George W. Bush) that "government should let people keep their money."  Ownership is a myth, after all, because we cannot know what is ours until the government's laws (including tax laws) have been put into effect, and differing versions of those laws -- each of which is as "natural" as any other -- will result in different people owning different things.

When I talk about the "baseline problem," I proceed from the same premise, but I am making a different point.  Because there are no pre-political arrangements of economic endowments, there can be no meaningful way to define "efficiency" in the orthodox economic sense.  Orthodox economics, after all, must have a baseline against which to compare the real-world outcomes that we actually experience.  If we cannot define what is efficient (even theoretically), then we cannot know what is inefficient, or even what is more efficient or less efficient.

I first discussed the baseline problem, without using that label, in 1998, in a draft of an article that was not ultimately published until 2005, as a book chapter under the title "Playing With Fire: Feminist Legal Theorists and the Tools of Economics."  Like Murphy and Nagel, I would never claim to have been the first to notice that there is no such thing as a no-government world.  Like them, I am working from that premise to undermine a political claim that has been cloaked in apolitical garb.  In my case, the claim in question is that there is a way to know that people's responses to a change in policy can meaningfully be condemned as inefficient.  (I also do not claim to be the only person ever to have made this basic claim about efficiency.  I do think that I am extending it in some new, and hopefully useful, ways.)

To summarize: Murphy and Nagel use the "no state of nature without government" premise to make a point about taxes, while I use that same premise to make a point about economic efficiency.

Because this post is already far too long, I will stop here.  In a future post (or, more likely, in a series of future posts), I plan to explore the question of whether there are meaningful ways, short of a fully specified baseline, to compare two policies (or two states of the world), and call one more efficient than the other.  That, it turns out, is a very difficult question.

For now, however, we can at least stop to enjoy the moment when a top-line conservative economist inadvertently revealed how meaningless is the notion of efficiency.  If the ACA is bad (or inefficient) because it will change people's decisions in the marketplace, then everything is bad/inefficient.  But then nothing is good or efficient, except whatever it is that might serve the political purpose of that economist and his comrades in Congress.