Testimony
by
Estelle James, Ph.D.
Senior Consultant to the National Center for Policy Analysis
House Committee on Financial Services
May 5, 2005
Over the past 20 years more than 30 countries, spread across Latin America, Eastern and Western Europe, Australia and Hong Kong, have adopted social security systems that include funded privately managed plans, usually based on personal accounts. Contributions to the accounts range from 2.5% to 12.5% of wages and they are projected to supply between 30% and 80% of total benefits.
In Latin America and Eastern and Central Europe the accounts were created by a carve-out from existing payroll taxes. In industrialized countries, such as Australia, Switzerland, Netherlands and Denmark, employers have long provided plans that covered about half the labor force on a voluntary basis. Governments decided it was important to cover the remaining half so they made employer-sponsored plans mandatory, as an add-on for employers that didn't already provide them. Although this option hasn't been much-discussed, this suggests one way that we could go in the US.
I am going to discuss how these 30 countries handled three issues-how to keep administrative costs low, how to control risk and protect low earners, and how to make payouts. I would like to stress two things: First, workers do not have free rein over the funds in the accounts. Instead, the accounts are tightly regulated and ownership rights are attenuated. The UK ran into trouble when it gave too much choice and too little regulation. Second, details matter. Seemingly small changes in rules can have a large impact on final outcomes. So you really need to look at dry details very closely.
Administrative costs
If a worker contributes to an account each year and pays an annual administrative fee that is 1% of the assets in the account, when he retires his accumulation and pension will be 20% less than it would be if there were no fee at all. Obviously, keeping costs and fees low is essential in order to get good value for money. Much criticism of personal account systems, such as that in Chile, has focused on its supposedly high administrative costs. Chile indeed had high costs in its first few years-start-up costs are always high-but currently they are 1.2% of assets per year and projected to be .7% of assets for full-career workers. This is lower than the average mutual fund IRA and 401k in the US.
However, I believe we should be able to do better still in a mandatory system, by exploiting economies of scale and eliminating marketing expenses. If we adopt measures such as competitive bidding for a limited number of asset managers, passive investment, and centralized record-keeping, I estimate that the expense ratio will be less than .3% or 30 basis points once the average account size exceeds $7000-that is, after 8-10 years of operations. This estimate is consistent with the Administration's plan.
However, if workers are given the right to opt out into a broader range of mutual funds once their accounts reach $5000, as some have suggested, the average account size in the basic system will never reach $7000 and costs will remain over .3% for everyone. This is a good example of how little details matter a lot.
Controlling risk and protection of low earners
We can never fully eliminate risk in financial markets but we can adopt measures that keep risk relatively low. Diversification across companies, sectors and even international diversification is a classic way to reduce volatility. Gradually reducing exposure to equities as retirement approaches, so workers are not hit with an unusually low stock market or interest rate on the date they convert to annuities is another important technique. In addition, every country that has a personal account system also has a minimum pension, most commonly 20-30% of the average wage. This is designed to protect workers from both financial market and labor market risk. So far, we do not have a minimum pension in our current system or in the proposed new system.
Payouts
Practically every country with personal accounts restricts payouts. Most European countries require annuitization, to ensure that workers will have a life-long income. In Latin America payouts must take the form of annuities or gradual withdrawals. In Chile, 2/3 of all retirees have annuitized. Lump sum withdrawals are not permitted unless the pension meets a high threshold, such as 70% replacement of the worker's own wage and 200% of the poverty line. This is much higher than the threshold proposed by the Administration, which allows lump sum withdrawals at 100% of the poverty line.
Some countries also require that annuities be indexed (to provide inflation insurance) and joint (to cover surviving spouses)–which is very important for women. In Latin America women are allowed to keep their own pension in addition to the joint annuity, so that married women who work in the market and contribute for many years are not penalized, as they are in this country. As a result, women's expected lifetime benefits relative to men's have increased in the new system.
Conclusion
In sum, the devil is in the details. Personal accounts can give us good or bad outcomes, depending on how we design them. The experience of other countries shows that if we constrain choice of investments and payouts and provide a pension floor, the redefined social security system will continue to provide lifelong income for the elderly in a cost-effective, low risk way.