The Impact of Social Security Reform on Savings and Capital Flows

The Impact of Social Security Reform on Savings and Capital Flows

November 15-16, 1996

I am certainly delighted to be here and pleased to see a lot of good friends. As a member of the Council on Foreign Relations, I am very pleased that the Council has taken the initiative to have this session and to be in the forefront of this very important issue. Social Security reform is a subject that I've personally been interested in for a very long time--literally decades--and during which I have done research on the impact of Social Security on savings. I believe that it has always been a major, perhaps the major, fiscal problem facing the United States, and I'm delighted that it's now beginning to get the attention and prospect of reform that a meeting like this indicates.

Recently, the National Bureau of Economic Research (nber) did a major study and had a conference on the feasibility of shifting in the United States from our current unfunded system to a system of funded individual IRA- and 401(k)-type accounts. As part of that, we looked at the experience in other countries in the same way that you all have during this conference. The thing that struck me, looking at what's been happening around the world--not just in Latin America, but in Australia and England--is that the process of transition has been very different in these countries. They start from different starting points. They have different political systems. But they have all moved to a system of much greater reliance on individual funded accounts with real capital accumulation.

My own interest in this project that we've been doing at the Bureau has been on the transition path--how we move from an unfunded system to a funded system over time. I'll draw on some of the results of that in my comments today about the international capital market aspects of this subject. The title of this conference includes the words "pension crisis" and the title of the current session is "the implication of the Pension Crisis for International Capital Flows." Calling it a crisis and thinking about the financial crunch that is clearly coming in the U.S. system may be helpful in terms of forcing political action, but I think the real problem is not financial crisis, it is a problem of a very inefficient system(a system that depresses savings and a system that imposes extremely high tax rates on individuals. Think about the current 12.5 percent tax rate and about the projections of how much higher it will have to be in the future. That to me is the real reason to move away from our current unfunded system into a system that can replace that with a much lower mandatory contribution rate and one that also leads to increased capital accumulation.

So, I want to talk about the subject of international capital flows with a slightly different focus-talking about the implication of the current Social Security financial situation and of the alternative Social Security financial situations for international capital flows. What I'll talk about specifically will be related to the United States, but those same comments about the United States can be translated, I think, to the context of other countries. I'll make a comment at the end on some of the slight differences that might be needed to make that translation.

As you know, the current situation of the U.S. Social Security plan is that it is in surplus, that tax collections and earnings in the trust fund exceed payouts by about 1 percent of gdp, but that that surplus is rapidly shrinking. The surplus is currently forecast to end in less than 15 years. Then the Social Security program will actually be adding to the size of the national budget deficit.

An alternative system, at least the one that I want to focus on, is a funded system with individual capital accounts. In principle, it could be a partial system or it could be a fully funded system. I'll come back to that in a couple of minutes. The question, though, that we in this panel have been asked to focus on is the impact of all of this on international capital flows. So, let's start with getting the basic issues straight.

In general, anything, whether it's a Social Security reform or it's a tax reform or it's just a change in domestic savings behavior that comes about spontaneously--anything that increases national saving, that is not offset by an equal increase in domestic investment in plant and equipment, housing, and inventories, i.e., anything that increases domestic savings, and which isn't offset by an increase in domestic investment, necessarily leads to an increased net capital outflow, or a reduction in the net capital inflow--a reduction in our current account deficit.

That is not a statement of economic theory. It is not even one of those empirical regularities that economists call facts. It is an accounting identity. It is necessarily true. If we increase our savings as a nation, and we don't increase our domestic investment, those funds have to go somewhere. They go abroad. So, we have a capital outflow.

Another way of seeing the same thing in terms of real resources is that if we increase our savings, that means we've cut back on our consumption. If those resources don't go into domestic investment, then those resources have been freed up to produce exports for the rest of the world, or to reduce our dependence on imports from the rest of the world. So, the basic fact is that if we increase our national savings from a Social Security reform or in some other way, and there isn't an equal increase in investment, then there will be a net capital outflow.

15 years. Then th

Let's go back now to the impact of Social Security and of Social Security reforms. The current situation, the unfunded Social Security system, undoubtedly reduces private savings. For most people in the United States, there is very little reason to do any saving at all because the Social Security benefits, at least as they have been paid in the past, replace a very large fraction of pre-retirement income. On an after-tax basis, Social Security replaces more than half of pre-retirement income for the typical employee. Benefits remain indexed so they keep up with inflation in retirement.

When we look at the actual data, we discover that most Americans reach retirement with about six months of financial assets--an absolutely astounding figure. When I tell it to people in other parts of the world, they say,"No, no, I didn't understand that; it can't be right." And I say that the average American retires with around $12,000. That's median financial assets for people in their early 60s--not at all irrational given the size of the Social Security program.

Well, you might say, "What about this surplus that the government is running in the Social Security program? Doesn't that offset the reduction in private savings?" At most, it offsets it a very little bit because that surplus is just one percent of gdp and rapidly declining. Moreover, when the government talks about a balanced budget in 2002, it takes that Social Security surplus into account, so that we get less of other deficit reduction effort because of the Social Security surplus. Therefore, the integrated budget of the government doesn't really improve because of the Social Security surplus, or it improves by less than the surplus.

So, there's no doubt in my mind that the net impact of our current Social Security system is a significant reduction(probably 3 or 4 percent of gdp in our national saving rate. If you relate that to the fact that the private saving rate in the United States is now an abysmal 5 percent of gdp, we're talking about almost halving our private saving rate and more than halving our national savings rate-- which is net of the government deficit--as a result of the Social Security program.

What does that mean? It means an increased capital inflow from the rest of the world and reductions in domestic investment. We are more dependent on foreign capital, and we have less investment here in the United States. What's that mix? How much of it is a change in capital flow and how much is a change in domestic investment? I will come back to that in a minute. First, I want to say something about the alternative to a pay-as-you-go system.

Although there are a variety of proposals that are being discussed, I think the common feature is movement to a funded system. What will that do to the national savings rate? That depends in part on whether it's a partial or a total replacement of the pay-as-you-go system--whether we continue to have the three-part system as the World Bank has advocated, or we move completely to a privatized system. It also depends on the rate of return that is earned by the contributions that individuals or their employers make to those privatized plans.

The calculations that I did as part of this nber study that we completed last summer dealt with the transition to a total replacement, i.e., a full substitution of a funded system for the existing unfunded system. That's not a political prediction. It's not an advocacy. It's just a benchmark for making the calculations.

To see what that means for savings, assume that benefits are to be maintained in a privatized system at the same level that is projected under our existing Social Security system, and that the real rate of return that is earned in the private retirement account is about 5.5 percent. That 5.5 percent has been the real rate of return over the long haul of the last several decades, in a mixture of debt and equity in the United States. It's also the after-corporate tax return that is earned on additions to the capital stock.

With these assumptions, the contributions to a privatized system that would eventually be required would be a little more than 5 percent of payroll. So, the roughly 20 percent of Social Security payroll tax that the actuaries now predict for the future could be replaced by a 5 percent of payroll mandatory contribution to a 401(k). That translates to about 3 percent of gdp. So, if we ask what is the impact of a mandatory individual retirement account-type Social Security reform on savings, the starting point is that it would initially add about 3 percent of gdp to our savings rate. How much of that would be offset by reductions in other savings? Very little, probably nothing. Why? Because most of the people covered by Social Security don't have any other savings to offset it against. Moreover, there is no change in the promised level of benefits in the experiment that I've just described, so there's no reason for them to reduce their other savings to the extent that they have any other savings.

So, I think we start with an increase of about 3 percent of gdp to the national savings rates. Well, now to come back to the question that I was asked to talk about, what is the impact of this on international capital flows?

If we add 3 percent to our national savings rate and there is no change at all in our domestic investment in plant and equipment, housing, and inventory, then that 3 percent will go into the net U.S. international capital outflow. We will send more money abroad. Given that we're currently a recipient of about 2 percent of gdp's worth of capital from the rest of the world, we'll go back to where the United States was before 1980, when instead of being a net importer of capital, we would become a net exporter of capital of about one percent of gdp.

I prefaced this estimate by saying if there is no change in investment. And the question is if we added 3 percent to our national savings rate, what would actually happen to investment? There has been a fair amount of research that I and others have done on this question. The evidence seems to say this: when your savings rate increases, your investment goes up as well. Not dollar for dollar, but, roughly speaking, for every extra dollar that we save domestically, our domestic investment goes up about 80 cents.

Savings tend to stay at home. High savings countries have high investments. Low savings countries have low investments. Look at the United States and Japan to take the extremes. There's Japan with a very high savings rate. They have a very high investment rate. Yes, they have a bit of a current account surplus--equal to a couple of percent of gdp. But they're starting with a 15-plus percent national savings rate. Come down to the United States. We have a national savings rate of about 4 percent of gdp. We have a little bit of a capital inflow. But our investment rate is very low. You look in between at the other countries of the oecd, or you can extend this to a broader set of countries, and you find that, roughly speaking, at the margin, an additional one percent of gdp that is saved leads to a capital outflow of about 0.2 of a percent and an increase in domestic investment of about 0.2 of a percent.

So, if as a result of this Social Security reform, we initially see an increase in our national saving rate of about 3 percent of gdp, which would be a massive increase relative to the very low level that we have today--if we have that 3 percent increase, about 80 percent of that 3 percent would stay at home--about 20 percent would move to offset our international current capital account flow. So about half of a percent of U.S. gdp would go into the international capital markets. With time, that would grow as the savings based on the new capital increases. Although these amounts are significant, they are not dramatic or in any sense an unsettling amount to put into the world capital markets.

What about other countries? What I have just said about the tendency of capital to stay at home in the United States, is also true for Japan. It's also true for Europe as a whole, although as Europe moves toward a single currency, there will be much more of a tendency for higher saving in one country to lead to a net capital outflow spread around within Europe. But for Europe as a whole, for the United States, and for Japan, I think, roughly speaking, increases in savings that will come about as a result of Social Security reform will tend to stay at home--about 80 percent will stay at home and about 20 percent will go abroad.

I'm going to be looking forward eagerly to hear what my fellow panelists have to say, and I look forward to questions as well.