On February 8 and 9 the Cato Institute held a conference, "Social Security: The Opportunity for Real Reform," in the F. A. Hayek Auditorium. Major speakers included Cato president Ed Crane, Harvard economist and former Council of Economic Advisers chairman Martin Feldstein, assistant to the president for economic policy Allan Hubbard, Social Security trustee Thomas Saving, and Cato senior fellow Jagadeesh Gokhale. Excerpts from their remarks follow.
Ed Crane: The Cato Institute has worked on Social Security since 1978, and we are delighted to see it as a major part of this administration's initiatives for the coming four years. From our standpoint, the essence of America is a respect for the dignity of the individual. And it seems axiomatic that the more control people have over their own lives, the more their dignity is enhanced.
In that regard, Social Security as it exists today comes up a bit short. In 1960 the Supreme Court ruled in Flemming v. Nestor that we as Americans have no right to the money we pay into Social Security. The Court said that Social Security is a social program of Congress and that what you get back at retirement is entirely up to the whims—I guess the word "whims" doesn't appear in the decision, so let's say entirely up to the discretion— of 535 politicians. That is why we at Cato think personal accounts are so important.
I think the three key arguments for personal accounts are ownership, wealth creation, and inheritability. But before I discuss those briefly, let me address the Social Security Trust Fund, because there is a lot of confusion about it.
People argue about what is going to happen in 2042 when the trust fund goes broke. But the reality is that the trust fund is already broke. There are no assets in the trust fund. When the Social Security Administration opens the lock box in 2018, there will be a lot of pieces of paper called special Treasury notes. And the Department of the Treasury, to redeem those notes, will need to increase taxes, increase borrowing, decrease Social Security benefits, or decrease other spending. But let's imagine that in 2018 we open the lock box and a moth flies out; there is nothing else inside. The options for the government would be precisely the same. To pay promised benefits, the government would need to increase taxes, increase borrowing, decrease Social Security benefits, or decrease other spending.
Therefore, I would suggest that the special Treasury notes have no value. And, in fact, I am not alone in that assessment. In the fiscal year 2000 budget of the Clinton administration, there is a whole section on how, in an economic sense, there are no real assets in the trust fund. So we should get beyond that and recognize that the cash flow will turn negative around 2018, and from that point forward the problem will get progressively worse.
I would not center this debate on the question of whether there is a crisis or not. What the opponents of personal accounts shy away from like a vampire from the cross is the issue of ownership.
Under a system with personal accounts, you would purchase real assets that you would own and that would be inheritable. And one of the most attractive aspects of the inheritability argument is recognition bonds that Rep. Sam Johnson's bill, which I think is unique among the other bills promoting personal accounts, provides for. As soon as you choose a personal account, you get a recognition bond that is your property and can be passed on to your heirs.
To me, it is outrageous that you can pay into the present system your entire working life and when you pass away the money goes poof and disappears. Where is the humanity in such a system? But with personal accounts—and particularly with recognition bonds—all of that money is inheritable, which is very important.
Martin Feldstein: We often hear that there is no need for structural reform. The basic facts, I think, are these: the Social Security system, because it is a pay-as-you-go system, is very sensitive to changes in demographics. The changing age structure is going to lead to very substantially increased taxes if benefits are going to be maintained. The tax for retirees' and survivors' benefits is now 10.6 percentage points of the 15.3 percent total payroll tax. The rest goes for disability and Medicare.
The Social Security actuaries tell us that that 10.6 percent will eventually have to rise to 16.6 percent, a 6 percentage point increase in the payroll tax. But even that is an understatement, because it does not take into account something that every public finance economist knows: that increasing marginal tax rates by 6 percentage points will shrink the tax base; people will work less and change their form of compensation.
So the actual tax increase, in order to meet that same financing goal and, moreover, to offset the loss of revenue to the personal income tax, would have to be even higher. The current combined payroll tax of 15.3 percent would have to rise to well over 22 percent. That would hurt the economy, and it would be a very substantially increased burden on middle- and lower-income households. Avoiding that tax increase completely would require cutting benefits by about a third.
Postponing reform until there is a crisis would mean an increase in taxes, because the political reality is that you cannot suddenly cut benefits. A failure to deal with the problem at the present time, when we have the time to lay the foundations, is really an agreement to have a tax increase in the future. If we want to avoid the unpleasant choice between unacceptable benefit cuts and a very damaging tax increase, we have to act now.
Can Social Security be fixed without structural reform? Well, without structural reform, without moving to an investment-based system of individual accounts, there really is no choice other than increases in taxes or reductions in benefits. One common tax increase proposal is to raise the amount of income that is taxable under the payroll tax.
Today, the payroll tax applies to the first $90,000 of income. And the proponents of this alternative say, well, why don't we just raise that to $120,000? I will tell you why we should not raise it to $120,000. That would create substantial disincentives to work for everybody whose marginal tax rate would jump immediately by 12 percentage points, for those between today's $90,000 and $120,000, and it would produce very little revenue.
With a colleague at the National Bureau of Economic Research last year, I looked at what would happen if we raised the payroll tax ceiling from $90,000 to $120,000. First of all, if there were no behavioral response at all, the amount of revenue that would be produced would be relatively small, about half a percent of payroll. So it would hardly make a dent in the shortfall.
But much more important is the fact that the higher marginal tax rates would shrink taxable income because people would work less and shift their compensation to fringe benefits and other untaxed forms. Even more important, because of the shrinking of taxable income, the personal income tax and the Medicare HI tax would bring in much less revenue.
Here are the numbers: If there were no behavioral response, then the proposed increase, or suggested increase, when we calculated it according to the rules as they were about a year ago, would have produced $19 billion. Taking into account behavior would shrink that to only about $5 billion. There would be about $11 billion of revenue lost to the personal income tax and the Medicare tax, plus about a $3 billion reduction in the amount of money collected by the payroll tax itself.
The result would be a big increase in marginal tax rates for an important group of people with incomes between $90,000 and $120,000 and virtually no additional revenue. That amounts to a backdoor way of shrinking the funds going into the personal income tax in order to build up the funds going into the payroll tax.
I think there is a serious problem. And without structural reform, we will have to see a big increase in tax rates or a big reduction in benefits.
Allan Hubbard: The president has made it very clear that he is not going to give people the opportunity to participate in risky personal accounts. His plan is based on the federal Thrift Savings Plan, where there are a limited number of index funds, stocks, high-grade corporate bonds, and Treasury bonds for you to invest in. There will also be a life-cycle fund, which will adjust your allocation among stocks and bonds to be appropriate for the number of years you have until retirement.
You are not going to be able to take the money out early. Just like Social Security today, you cannot take the money out early. You cannot borrow against it. When you retire, you will be able to take a lump-sum payment out of your personal account. But you will have to leave in enough, when combined with what the Social Security Administration owes you, to ensure that you will never go below the poverty line. The president believes very strongly that there has got to be a comprehensive reform to fix Social Security permanently. And he also believes that there is no comprehensive fix that does not include personal accounts. Without the personal accounts, any fix is going to be unfair to younger generations.
There are enormous benefits to personal accounts. Number one, a personal account is going to be something you own and you control. The government cannot ever take it away. No matter what Congress decides, it cannot change the benefit structure. That personal account is yours forever. It is a nest egg for retirement. And if by chance you were to die before retirement, it is something that would go to your estate and would be passed on to your heirs.
The current system does not give a very good return on your Social Security taxes. The actuaries have found that, on average, you will get a much higher return from a personal retirement account than you will from money going into the Social Security system.
Some people claim that President Bush's proposed personal retirement accounts actually benefit the federal government more than the account holder by providing a clawback. That drives me absolutely crazy. There is no clawback. I do not know how anyone could have ever gotten that idea.
When you get a high return on your personal account and then the government takes a piece of it back, it's called a clawback. That's not how the president's plan works. The government is not allowed to touch anything in your personal account. No matter how successful you are, it is yours forever.
Thomas Saving: Is the crisis real? In considering that question, we need to think about several important dates. One is the date of the first deficit, 2018. Another is the date of peak contributions to federal revenues, 2008. Each year after 2008, Social Security's contributions to Congress's ability to spend will decline. The Treasury is going to feel the pinch not in 2018 but in 2009.
Another important date is the date when the share of federal tax revenues required to pay benefits reaches 5 percent. That is 2021—just three years after the system starts running deficits. At that point we will already be transferring 5 percent of federal income tax revenues to Social Security.
Only twice have we had to transfer significant amounts of the federal income tax revenue to Social Security—in 1978 and 1982. The maximum we have ever transferred is 4.5 percent of federal income tax revenues. So three years after the system turns negative, we are already going to be transferring more than we have ever transferred.
What did we do the two times we transferred large amounts? We changed the system. In 1983 we raised taxes by accelerating scheduled tax increases and reduced benefits by raising the retirement age for future generations and by taxing benefits.
What about the Social Security Trust Fund? Keep in mind that the trust fund provides no revenue to the Treasury, and the benefits have to be paid from Treasury revenues. What is the trust fund then? In a sense, the way current law reads, it is the authorization to pay benefits. In effect, it is as if I said to all of you, "I authorize you all to take two weeks and go to Paris if you can find the money." If you don't have the money, authorization does not mean anything.
Only the cash flows between the Treasury and Social Security really matter. And those cash flows have nothing to do with the trust fund. So when we have to start transferring money to pay benefits, we have to find that money somewhere. The trust fund will not help us.
Once we recognize that the trust fund cannot provide anything to pay these transfers, then the real transfer that the current working generation has to make to the present retired generation, and future working generations to future retired generations, is $13 trillion. That is how much money they have to come up with, because the trust fund does not have any real money in it.
Jagadeesh Gokhale: Over the next 75 years, the Social Security system has a projected financial shortfall with a present value of $3.7 trillion. That means that today we would have to put $3.7 trillion in the bank, and earn interest, in order to meet our obligations over the next 75 years. But it's worse than that. The demographic imbalance is permanent and will continue to worsen over time. So we have to look further into the future, even further than 75 years, to see the full extent of the problem. In fact, the long-term shortfall in Social Security is about $12 trillion. That is, we would need $12 trillion in the bank today, with interest, in order to permanently restore Social Security to sustainability.
If we don't have $12 trillion in the bank today, then we must raise an equivalent amount of resources by changing Social Security's current tax and benefit policies, because those policies are unsustainable. We have to either increase taxes or cut the obligations that those taxes finance in terms of Social Security benefits.
Now, the tough question becomes: Who bears the burden of these fiscal adjustments? How much will each birth cohort have to pay? Most reform proposals set forth the principle that those who are retired already or close to retirement—roughly speaking, 55 or older—are not going to bear the burden of any policy changes. So the entire burden of financing the $3.7 trillion shortfall over 75 years (or a $12 trillion shortfall in perpetuity) must be borne by those who are younger than 55 today and by future generations.
You have only two variables: taxes and benefits. One way to deal with the problem is to improve the quality of benefits. How could we do that? We might allow a worker to invest a portion of his payroll taxes in personal accounts, which would be invested in private market securities. In exchange, the government's future benefit obligations to that worker will be scaled back.
How does that improve the quality of benefits? Well, we know that scheduled benefits today for Social Security are more than what is payable today under today's tax laws. And in addition, polls show that most young folks do not really expect to receive the scheduled benefits. Many people actually say that they do not expect to receive anything from Social Security when they retire, because they doubt that the program is actually going to be around then.
So reducing scheduled benefits in exchange for the ability to take a part of the payroll tax and invest it in personal accounts might be a good bargain for some young folks. In addition, the quality of benefits would be better with personal accounts because personal accounts would have additional features. For example, people would have greater flexibility in withdrawing assets from their personal accounts and would have the option to bequeath their accounts to their heirs.
Of course, there is a further problem. We are talking about adjustments for the people younger than 55. But those older than 55 still have to be paid their benefits. If fewer payroll tax revenues are coming into the Social Security system, the government has to raise money from somewhere else to make up the shortfall in order to pay current retirees their benefits.
The standard assumption is that the government will just borrow the difference. So if, say, 4 percentage points of payroll taxes are invested in personal accounts for those who are 55 or younger, the government will just make up that shortfall in revenues by issuing additional government debt. The increase in the explicit debt of the government is what people call the transition cost.
I'm not sure the label "transition cost" is appropriate. Normally, when the government incurs debt, it means that the government is mopping up resources that the private sector could have used for investment in private ventures. But this addition to the government's explicit debt is different. It is not going to mop up private resources because, at the same time, the diversion of payroll taxes into personal accounts invested in private markets adds to private resources. So on the one hand, you are putting money into the private economy. On the other hand, you are borrowing money. So this is not necessarily a cost. It might be a neutral transaction. It might even be better than neutral, because the government might cut other wasteful spending programs to finance the transition. In any event, the government is not necessarily draining resources from the economy.
The other way to look at it: although the government's short-term debt is going up, it is doing so in exchange for reducing future obligations. Either way you look at it, it is a mistake to label the transition financing for Social Security reform a "cost." A better label would be "transition investment."