Do Bush's Accounts Move Toward Liberty?

The Colorado Freedom Report:  An independent journal of politics and culture.

The Colorado Freedom

Do Bush's Accounts Move Toward Liberty?

by Eyal Mozes and Ari Armstrong

On January 7 Eyal Mozes posted a reply to the e-mail group We the Living about my article concerning Ayn Rand and Social Security. He granted permission to reproduce his comments at Following his comments, I offer a response. While neither of us convinced the other, the exchange proved helpful for sorting out the various issues. -- Ari Armstrong, January 12, 2005

Eyal Mozes

Ari Armstrong criticizes Bush's plan for partially replacing the social security system by private accounts, and criticizes the Cato Institute and some ARI writers for supporting Bush's plan. I agree with Ari on many points, but overall I disagree with his conclusion that the plan is not a step in the direction of liberty at all and that Objectivists should not support it.

Some points on which Ari is clearly right are: that mandatory private accounts will not solve all the problems with the current system; that the only way to solve all the problems with the current system is to abolish it completely, probably in several stages through a phase-out plan; that any free-market advocates who support the private-accounts plan should publicly and explicitly endorse abolition of the system and make clear they support the plan only as a partial improvement over the current situation; and that the Cato Institute, and some ARI writers, have been remiss in not making this clear.

Ari claims that the private-accounts plan is in fact a combination of two totally separate and unconnected proposals: allowing workers a partial opt-out of social security, and setting up a system of mandatory private retirement accounts. On this, I disagree. As long as we have income taxes that make it difficult to save for retirement, the only way individuals can effectively save for retirement is through special accounts that allow savings to grow tax-free. We have some provision for this today with 401(k) plans and IRA accounts; but any plan to eliminate or reduce the current social security system, whether through phase-out or opt-out, since it will mean individuals can no longer expect to receive social security benefits after retirement, will have to be accompanied by some additional provision for individual saving for retirement. I do completely agree with Ari that there's no justification for making such new savings provisions mandatory; the most reasonable solution would be a phase-out of social security, along the lines that Ari has suggested, combined with some provision for additional voluntary saving, e.g. through a drastic increase in personal IRA contribution limits. But the point is, some form of provision for private retirement accounts has to be an inseparable part of social security reform; the link between the two is not arbitrary as Ari claims.

Given that Bush's private-accounts plan clearly does not solve all the problems with the present system, there are three main questions we need to answer in order to evaluate it:

1. Is it a significant step towards solving at least some of the problems with the system?

2. Is it likely to lead to further reform towards greater liberty in the future, or is it likely to block such further reform?

3. Will it be in any respect worse than the current system, causing new problems and new impingements on freedom?

To answer question 1, we should identify that the current social security system has two basic problems:

a. The coercion problem. The system is a coercive, paternalistic system, forcibly taking individual's earnings, preventing them from using it according to their own judgment, and employing this coercion allegedly for their own benefit.

b. The pyramid-scheme problem. The system is a pyramid scheme (sometimes referred to as a "ponzi scheme"), in which current "contributors" are promised future benefits not from any productive investment but from the contributions of future "contributors."

Note that these are two separate problems. The ponzi-scheme problems has serious consequences in addition to its coercive nature. It is the cause of the financial unsustainability of the system. It is also the cause of a grave injustice towards those who die young; those people have been forced to "contribute" to the system during their working years as much as anyone else, but they and their heirs are deprived of the promised benefits.

A system of mandatory private accounts will be no better and no worse than the current system as far as the coercion problem. But it will help in solving the pyramid-scheme problem (a complete conversion to mandatory private accounts would completely solve the pyramid-scheme problem; a partial conversion, as Bush is proposing, would partially solve the problem). For that reason, I think it would be a significant, though incomplete, step towards greater liberty and towards solving the problems of the current system.

Regarding question 2, Ari claims that Bush's plan threatens to block meaningful reform in the future. I disagree. On the contrary, I think that if Bush's plan is implemented, further reform would become much easier. The biggest political hurdle, that must be faced by any attempt to reform social security, is the problem pointed out by George Bernard Shaw: "A government that robs Peter to pay Paul can always depend on the support of Paul." In this case, younger workers are Peter, retirees and people close to retirement are Paul, and any attempt at reform must overcome opposition from a very large number of Pauls whose ability to receive benefits at Peter's expense is threatened. Private mandatory accounts will overcome this hurdle; once they are in place, proposals for further reform, by making contributions to these accounts voluntary rather than mandatory, and by easing regulation on these accounts, will not be a threat to anyone's benefits, and will thus be accepted much more easily.

On question 3, Ari notes that because the private accounts will include stock-market investments but will be subjected to heavy regulation, their investment decisions would be subject to political pressure, perhaps even made by government officials. This would be a new problem, a problem that does not exist in the current system and could be created by private accounts. Because the total amount of money in social-security private accounts is likely to quickly become very large, it could become a significant portion of the total money invested in the stock market, and having the investment of this money directed by government officials would create a lot of opportunities for political manipulation of the financial markets and of corporations.

I agree with Ari that this is a serious concern. But I don't think it is unavoidable. From what I have read about the Chilean system, it looks like they successfully avoided this problem (I don't claim to be an expert on the Chilean system, and I may have the wrong impression here, but this is my impression from what I have read). The mandatory retirement accounts are managed by private companies. Workers are required to put a certain portion of their payroll in these accounts, but they have a genuine choice in choosing among the companies. The accounts are subjected to heavy regulations restricting what percentage of the funds can be invested in stocks, and the managing companies are not allowed to engage in any other investment or banking business outside of managing retirement accounts; these regulations certainly make the system less than ideal; but as far as deciding what stocks to invest in, the decisions are made privately, no different from those of any mutual fund manager. This is far from a fully free system, but it is clearly more free than the US social security system, and does avoid the problem of government-directed investment.

In contrast, the plan proposed by Cato (in the paper written by Michael Tanner) would not avoid this problem. Cato propose a three-tier system, and in tier II (which will contain most of the accounts) individuals will have a choice among three funds; each fund will be invested in a diversified portfolio of stocks and bonds, each with a different ratio of stocks to bonds. This means presumably that the social security administration will appoint the managers of each of these three funds, who will then control the investment decisions of most of the social security private accounts. It is likely that the total of these accounts would quickly become very large, thus creating a lot of opportunities for these government-appointed fund managers to exert control over the financial markets and over corporations through their investment decisions. I find it shocking that a free-market organization like Cato would propose such a plan.

Roger Bissel says that a private accounts plan would amount to a fascist program. Regarding Cato's proposal, I agree. The defining characteristic of fascism is that property is nominally private but decisions about its use are made by government officials; Cato's plan - accounts which would be nominally owned by individuals but invested according to decisions made by the government-appointed managers of three funds - clearly fits this definition. But regarding the Chilean system, I think Roger's characterization is wrong. Under the Chilean system, individuals are required to put a certain portion of their payroll into the mandatory retirement accounts, but have a genuine choice among companies to manage their accounts, and the investment decisions in these accounts are subjected to a lot of regulation but are still made privately; this is not fascism, it is a government-hampered market system, and it is clearly preferable to the socialist system we have in the US today.

At this point we have no way of knowing whether Bush's plan will be more similar to the Cato plan or to the Chilean system; so far he and his advisers have not come up with any details about how the private accounts will be managed. If he comes up with a plan in which all private accounts are required to be in one of a designated small number of funds, then his plan, by creating government-directed investment in the stock market, may create more problems than it would solve.

However, if Bush comes up with a plan for private accounts managed by private companies, with individuals having a genuine choice in choosing the company to manage their account, and with the companies, even though subject to regulation, still making their own investment decisions and not having these decisions dictated by regulators; then I think Objectivists, and other free-market advocates, should enthusiastically support the plan. In supporting the plan, we should denounce its coercive aspects, explicitly endorse full freedom, and make clear that we support the plan only as an improvement over the current system; but we should also recognize that such a plan will indeed be a significant step towards liberty compared to the current system, and will make it easier to achieve further reform towards greater liberty in the future.

Ari Armstrong

Mozes's excellent criticism points to the difficulty of trying to figure out the consequences of proposals by pragmatic Republicans in a severely mixed economy. Untangling the interventionist mess left by the soft socialists of the early 20th Century is no easy task.

What I most appreciate about Mozes's reply is his concern for espousing the correct (free market) principles even as we adopt short-range incremental policies toward greater liberty. If we keep the context of economic liberty, we are more likely to ensure incremental reforms are relatively good and more amenable to positive change in the future. Thus, while I continue to disagree with Mozes on a number of points, and I doubt I'll be able to bring him completely to my perspective, I acknowledge that the way he promotes the accounts is basically good and a useful way to pursue the debate.

Mozes's most powerful argument is that mandatory accounts would be easier to change in the future than is the Social Security system, which is a scheme purely to transfer wealth. At least the accounts would be nominally owned by those who fund the accounts, even if use of that money is severely restricted by the national government. Making the accounts less restrictive or noncompulsory would not impact the promised welfare benefits of any other party. I sincerely hope that, if we end up with mandatory, regulated accounts, the future reforms Mozes envisions will come to pass.

However, the mandatory, regulated accounts would also bring with them new risks, the gravity of which Mozes may not sufficiently recognize. If one possibility is to make the accounts less restrictive in the future, another possibility is to make the accounts more restrictive and more expansive. How this will go will depend upon the political climate. I regard it as a dangerous bet.

Even the Chilean system praised by Mozes is relatively restrictive. (Interestingly, the Heritage Foundation now regards Chile as more economically free than the United States. However, the analysis seems to assume, as the comments concerning the United Kingdom suggest, that converting Social Security to a system of mandatory, regulated accounts is automatically a sign of liberty, which I think begs the question.) As Piñera discusses in his article about Chile, the companies that manage the pensions "can engage in no other activities and are subject to strict supervision by a government agency." He continues, "The law encourages a diversified portfolio, with no obligation to invest in government bonds or any other security."

Unfortunately, politicians would constantly be tempted to tinker with the regulations. Various interest groups will want favored (or disfavored) status for different types of investments. I fear the U.S. plan will mandate "investment" in government bonds, which would be nonproductive. (I also have a moral problem with purchasing government bonds.) Leftist groups will of course favor companies that are "environmentally friendly" or that employ labor practices the left desires. Even if additional restrictions are not adopted, the constant threat of such will pressure U.S. companies to play ball with the national government.

A system of mandatory, regulated accounts automatically disallows most types of investment. Workers cannot spend that portion of their income on things such as actual gold, real estate, personal businesses, stocks of consumables, etc. Thus, mandatory, regulated accounts skew the entire economy toward business entities with the size and desire to issue stocks. The mandatory accounts further entrench the severe national controls over the stock market.

Maybe Mozes is correct that restrictions of mandatory, regulated accounts would eventually become less severe. But I think it's equally likely that the restrictions will become more severe. That's a frightening risk.

Another risk is that national politicians would come up with new ways to force us to spend our money. The fundamental pretext for Social Security is that elderly people can't be allowed to slip into poverty. As bad as such a pretext is, the number of impoverished persons is limited. On the other hand, the fundamental pretext for mandatory, regulated accounts is that people are too stupid and/or irresponsible to use their income prudently. (Keep in mind that any opt-out plan could work as well by simply letting workers use the residual funds however they want.) Under such a pretext, there is no logical limit to the amount of money national politicians should control. If it's good to force us to invest some of our money for retirement, isn't it also good to force us to allocate additional money for education, medical expenses, groceries, transportation, and so on? The logical conclusion to the assumption that workers are fundamentally stupid and/or irresponsible, while national politicians are wise and benevolent rulers, is that 100% of our income should be directed by national mandates.

Mozes argues, "A system of mandatory private accounts will be no better and no worse than the current system as far as the coercion problem." Unfortunately, that's not quite right. Piñera notes that Chile's system required deficit spending, "so the cost was shared with future generations" (how generous). In the U.S., the switch to mandatory, regulated accounts is expected to require at least a trillion dollars of new deficit spending (i.e., future tax hikes). In an important sense, then, the national government would control an even larger portion of our money, at least for several decades: the 12.4% that would be split between Social Security and mandatory, regulated accounts, plus the additional amount of money required for deficit spending. Mozes might reply that the magnitude of control is less over that portion of the money directed toward mandatory, regulated accounts, but that seems ambiguous. (I've argued elsewhere that the "transition costs" should not be weighed against future promised benefits.)

Mozes also argues that the reform of Social Security requires more tax breaks for investment. However, his argument justifies only tax breaks for voluntary investments; it does not justify mandatory investments.

Any specific tax break makes the tax less neutral; i.e., more likely to influence behavior away from what it would be on a free market. Thus, while I sympathize with the idea of tax breaks for investment, I'm not sure they're a good idea. Today, the tax code is used as a tool for social engineering, which is terrible. So long as any tax on income exists, it should be as simple and low as possible. Also, only certain types of investment are given special tax breaks, which skews the economy toward those investments and away from others. Thus, I'm not convinced that investment (or anything else) should receive any special tax consideration. Regardless, the matter is not relevant to whether we should have mandatory, regulated accounts.

The main consideration, though, is what proposal we as advocates of the free market should actively support. The phase-out plan is obviously superior. It unambiguously reduces national control over our money. It can be implemented such that the Social Security tax is continuously lowered over time, without any increase in any other tax. I remain unconvinced that replacing part or all of Social Security with mandatory, regulated accounts would be a step toward liberty. Even if we assume they are, though, they are a small step compared with the massive amount of economic liberty that would be achieved through a phase-out plan. Thus, I don't see why any free-market advocate would spend even a second of time pushing for mandatory, regulated accounts, when that same effort could be spent promoting a phase-out plan. Nevertheless, the right principles can make up for a lot of strategic error, and so even if Mozes promotes mandatory, regulated accounts, I'm sure he'll do so in such a way as to make the attainment of economic liberty more likely.

For additional analysis, please see Social Security: Collected links.

The Colorado Freedom