Thursday, August 13, 2015

The U.S. and Australia Are the Same, Only Different (Retirement Edition)

by Neil H. Buchanan

In my Verdict column earlier this week, I ran through a few of the most common arguments that conservative critics of Social Security repeat ad nauseum, showing each of those arguments to be based on nothing more than an inability to understanding basic accounting.  I then used the most obviously false of those arguments -- that Social Security is a Ponzi scheme -- to frame Tuesday's Dorf on Law post, in which I explained how and why private savings accounts are no more "real" than Social Security's finances, including the much-misunderstood retirement trust fund.

People imagine that banks hold piles of money, whereas Social Security supposedly spends its money right away.  In fact, both banks and Social Security send their money right back into the financial system as soon as they receive it, yet both are able to keep the promises that they are making.  The argument should not be about whether one set of promises is more real than the other, because it is legal and ultimately political commitments that underlie the ability of all financial promises to be kept.  If those commitments change, then of course outcomes could change, too.

I should note that, in nearly all of my writings to date, I have taken as a given the claim that Social Security's promised benefits come with a legal asterisk, that is, that the "promise" that Social Security has made goes like this: "Based on your earnings trajectory, you will receive $X per month in benefits when you retire (setting aside early or late retirement).  BUT, if the trust fund ever runs to zero, your benefit will be reduced by y%."  That is the asterisk that Social Security includes on its official forms (although it has been impossible for me to find out when or why they started to do that), and the annual trustees' report provides the latest best guess of whether and when the trust fund will reach zero, and if so, the value of y.  (The trustees' most recent preferred estimate says yes, it will run to zero in 2034, and y = 21%.)  For reasons that I will explain in a future column and/or post, even that is not necessarily true.  That is, it is an open legal question of whether there will be any reduction in benefits, even if the trust fund goes to zero.  This means that it might not end up being true that low-information people will be unpleasantly surprised by a one-time cut in benefits.

That explanation, however, will have to await another day.  Today, I published a new Verdict column, in which I discuss recent proposals by progressives to go on the offensive regarding Social Security.  That is, rather than remaining in a reactive, defensive crouch, simply responding to baseless attacks on Social Security in an effort to maintain the status quo, some of the highest profile progressives in Congress (currently led, of course, by Senator Elizabeth Warren) and the leading left-leaning policy think-tanks (especially the Economic Policy Institute) have started to demand an economically progressive increase in Social Security benefits.  This is a fight that was started by now-former Senator Tom Harkin before he retired.

In today's column, I do not go through any of the details of the Harkin proposal, and I will not do so here.  (It is a safe bet, however, that I will get to that soon enough.)  Instead, I simply describe the Harkin proposal as increasing benefits for lower- and middle-income retirees, fully financed by progressive revenue increases.  In fact, the proposal also includes financing provisions that would make the "21% cut in 2034" thing moot, so that it deals with any lingering concerns about the current system and then adds a paid-for progressive expansion of benefits.

My column runs through the obvious reasons that such a plan is needed.  More and more people rely on Social Security, mostly because of the rise of inequality (which, from workers' standpoints, has simply meant stagnant real wages since about 1980 or so), which prevents workers from saving enough money in defined-contribution savings plans.  The Social Security system is quite modest, currently providing retirement benefits averaging $16,000 per year ($1,333 or so per month).  This proposal thus provides the late-in-life component to complement the "America needs a raise" movement.

The centerpiece of today's column, however, is the question of whether it is too politically risky even to offer to open up the Social Security system to a hostile Congress.  Perhaps the best that can be done is to continue to play defense.  I ultimately come down in favor of giving it a try now, but I concede that it is a reasonable argument either way.  Interested readers can read more in the column.  For the balance of this post, however, I want to discuss the other major issue that I discussed in today's column, which is whether it would make sense to expand retirement savings, through any of a variety of policy levers.

As I point out in the column, the idea of giving people more incentives to save will not solve the immediate problem facing Social Security-reliant retirees.  For that matter, any such plan would really begin to make a difference only for people who are a few decades away from retirement.  Even so, imagine that the Harkin proposal goes nowhere, for any of a number or reasons, but Congress considers a long-term expansion of retirement savings incentives.  Or imagine that we decide both to increase Social Security benefits right away and also to put more savings incentives in place, in the hope that the future private savings might allow a future Congress to peel back some of today's increases.

Either way, we are left with the dilemma of how to structure a plan to enhance the private retirement savings that are currently driven through the various 401(k)-type programs.  In a pair of Dorf on Law posts earlier this summer (here and here), I described the fundamentals of such "neoliberal" approaches to retirement security.  As I noted, any serious effort to expand private retirement saving has to confront the many cognitive biases that distort the decisions made by even the most savvy people.  (In the second of those blog posts, I noted a Harvard economics professor's description of his own far-less-than-optimal behavior when it comes to planning for his retirement.)  My argument there was, in one sense, reductio ad absurdum; that is, I argued that doing everything necessary to fight the various cognitive biases would end up creating a system that would look very much like Social Security, but with higher administrative costs (and, by the way, without the progressivity).

Imagine, however, that we are for some reason committed to using private savings accounts to enhance retirement security.  What would that actually look like?  An article (forthcoming in the Indiana Law Journal) by Marquette Law School Professor Paul M. Secunda provides an interesting positive description of such a system, as well as a normative case for adopting that system.

Professor Secunda first describes why the current 401(k)-led system has been a disaster.  The bulk of his analysis describes the behavioral economic case against relying on people to make long-term decisions in a complicated environment.  I have expressed serious doubts (here and here) about the usefulness of so-called Behavioral Law and Economics (BLE), but I have always acknowledged that BLE at least tries to confront real-world phenomena that standard economic theory ignores.

In any event, Professor Secunda lays out a very convincing case for what he calls "paternalistic workplace retirement plans."  They are paternalistic precisely because even some Harvard Economics professors cannot fend for themselves in a wide open financial system.  All kinds of restrictions are necessary.  Professor Secunda then points out that such a restrictive model actually exists: Australia's "Superannuation Guarantee," which essentially gives workers zero control over how their savings are invested in the financial markets.  Imagine a nationwide system in which Social Security payroll taxes go into a single mutual fund, and the best financial managers invest the funds with relatively low management fees.

[Update: It turns out that my description of the Australian system in the paragraph above might best be described as a rose-colored-glasses version of reality.  I have written a short update here, explaining how some additional information changes my assessment of the Australian system, but in a way that supports my larger argument in this post.  Certainly, I no longer can confidently stand by my positive statements in the paragraph immediately below this one.]

The evidence indicates that the Aussie system works rather well.  Professor Secunda, for his part, concludes from his investigation that we should use the Australian approach to reform our current 401(k)-led world, to the benefit of workers.  I wholeheartedly agree.  Opponents of Social Security, however, could point to Australia's system as proof that private accounts could do all of the work that Social Security does here in the U.S.  Is that the right lesson?

At best, there is a possible argument that, if we were starting from scratch, we could design a "fully funded" retirement system that is functionally equivalent to a pay-as-you-go system.  But we are not starting with such a system, and as I argue in my Verdict column today, even a proposed expansion of Social Security needs to be pay-as-you-go.  Moreover, Professor Secunda's argument supports my point that the only safe alternative to Social Security would be a backdoor Social Security system.

Even if we ignore immediate needs and think only about making changes that would have good effects decades from now, therefore, there is no good case to change the current Social Security system.  And if we want to enhance future retirement security, simply adding onto the current system would have the lowest administrative costs, and it could accomplish everything that an Aussie-style system would achieve.  The current system in the U.S. is readily expandable.  The only question is whether the people who oppose the system on ideological grounds will succeed in destroying it, or at least will end up forcing us to adopt an add-on system that squanders Social Security's systemic advantages.

No comments: