Summary, September SGE Monthly Luncheon
“Building Automatic Solvency into U.S. Social Security: Insights from Sweden & Germany”
Summary of Remarks by James Capretta
Fellow, Ethics and Public Policy Center
By Sviatlana Francis, Group of ThirtyMr. Capretta delivered remarks on the social security reforms in Sweden and Germany. He shared his views how these countries’ experience can help address the issues of the social security reform in the U.S. The irony of the topic is that we can in fact learn from the welfare-oriented economies in Europe. No major social reforms have been undertaken in the U.S. since 1983, while Europe has introduced a few reforms since. We can gain some useful insights in how to approach this topic and how much the reforms have changed the system. Sweden and Germany provide particularly interesting examples in this regard.
Sweden experienced a deep economic recession in the early 1990s. The new center-right coalition put state pension reform on top of its political agenda and it was able to reach an agreement with the Social Democratic party. The new pension system, adopted in 1998, combined two approaches: individual accounts, as supported by the center right, and pay-as-you-go, as supported by the Social Democrats. The result is called “notional defined contributions” which look like personal accounts, but without advance funding. As a result, the overall social security system became tighter, but thanks to additional welfare protection mechanisms guaranteed by the Swedish government, the system remained loose enough to allow implementing the reform.
Sweden's goal was to maintain a fixed contribution rate of sixteen percent of wages, but in reality a total payroll tax is higher—18.5 percent. The workers are required to set aside additional 2.5 percent of wages as the additional personal contributions of the pay-as-you-go system. However, for political purposes, the tax remains 16 percent, at least on paper.
Sweden has preserved its overall entitlement program but it changed the way it is calculated. Workers’ 16 percent payroll contributions are treated like investments, comprising “notional defined contributions” that in reality are used to finance current retiree's benefits. The contributions on the pay-as-you-go basis are credited back with the rate of return equal to the growth in average wages when the taxpayers retire. Thus, one is still entitled to what he pays into the system, but it depends on two factors: the contributions' rate of return and the contributor's retirement age. Such system insures that the entitlement is directly connected to the demographics, as well as economic dynamics.
Sweden introduced a new tool—annuity divisor, similar to the mechanism used by the insurance companies, to calculate the pension payment from the notional fund. The annuity divisor is an adjusted estimate of the expected period of pension-drawing based on the birth year and the average life expectancy in accordance with the demographic trends and medical advances. If one is born in 1930, the annuity divisor is 14.8 and it has no effect on the pension. If one is born in 1940, the annuity divisor is 15.8 and it minimizes the pension by 5 percent. The retiree has an option of retiring at 65, which is the retirement age in Sweden, and taking the reduced pension or working an extra year and receiving the full pension amount, as if he they should work longer to contribute to the system accordingly and qualify for a full pension amount. This system was difficult to introduce initially, but once it was implemented in 1998, it continued working.
However, the annuity divisor was not the only mechanism, necessary for smooth functioning of the system. In 2001, Sweden introduced an “automatic balance mechanism.” It treats the social security payment system as a balance sheet, i.e. the difference between assets and liabilities. Contributions to the system comprise assets, while liabilities are pensions, calculated as the contributions' payback with the presumed rate of return. By default, the latter are calculated based on the per capita income. However, the demographic trends (longer life expectancy, lower birth rates) show that the system cannot afford high rates of return. Thus, if the balance ratio (assets over present value of pension liabilities) is lower than one, the system takes a haircut out of the pension pool; if the balance ratio is higher than one, full indexing occurs.
In order to calculate “the assets,” expected contributions are multiplied by the so-called expected turnover duration. The latter indicates the period of time, calculated as the difference between earnings-weighted average age of the current taxpayers and the pension-weighted average age of the pension recipients. It is currently estimated at 32 years, i.e., in pay-as-you-go there are 32 years, on average, between the time workers pay into the system and earn an entitlement and the time that entitlement will be liquidated with a pension payout. Current demographics and income levels show that the balance ratio might be below one (i.e. liabilities exceed assets) and the automatic adjustments may come into play soon. If the future work force contracts, the turnover duration shrinks, decreasing the pension pool further. Many other factors can affect the amount of the pension pool, such as retirement age, immigration rates, birth rates, life expectancy, etc.
Germany has a very generous social security scheme, but it taxes at a high tax rate to pay for it: a 22 percent payroll tax. Schroeder's government introduced two sets of reforms in 2001 and 2004. The first commission introduced voluntary accounts, which, however, did not become very popular. The second commission looked at the so-called “sustainability factor” when calculating a new benefit formula. It compared the numbers of people who used to contribute to the system with the projected numbers of the taxpayers. Currently they adjust the pay-off amounts every year, taking into account a number of economic factors.
The main lesson that we can learn from both of these examples is that changing demographics demand the reform of the social security. The current system is neither balanced nor stable, but implementing social security reforms allows making it more consistent and more stable. We need to build in automatic adjustment mechanisms, which will reduce the political pressure on each government to plan another system overhaul. Due to the current dynamics, it is unavoidable to experience plummeting in the pension pool but the system can prepare for it by changing its indexes and adjusting to the current demographics.
During the Q&A period, Mr. Capretta reiterated that the social security reform would affect all generations, young and those close to the retirement age. There is a difficulty of bridging the gap between the expected rate of return and the realistic pension amount, but cutting entitlements is unavoidable. It might be possible to honor previous commitments by creating notional accounts. Mr. Capretta admitted that Swedish model is possible to a large degree due to a wealth of other welfare benefits available to the population, as well additional phase-in/phase-out mechanisms that permit the government to monitor the system.