Implications of the Pension Crisis for Capital Flows

Robert D. Hormats

Council on Foreign Relations

November 15-16, 1996

I have been asked to discuss the implications of the pension crisis for capital flows.

The coming clash between demographics and government commitments to provide rising pension benefits will force virtually all countries to reform their national and in some cases their private pension systems. They must do so in order to hold down payroll taxes for the relatively smaller portion of their citizens who will be in the active workforce paying into the system and to ensure a comfortable old age for growing numbers of retirees. This means, inter alia: 1) funding a larger part of social security programs, 2) investing in more productive domestic assets to earn a higher return and promote investment to sustain economic growth, and, 3) over time, diversifying to invest a growing portion of their reserves internationally.

Let me begin with a few words of introduction.

The "corporate privatization revolution" began largely in the industrialized world; its intellectual impetus initially came from the Thatcher Government in Britain. In contrast, the "pension privatization revolution" has, for the most part, received its initial intellectual impetus from newly industrialized and developing countries--although to be sure it encompasses a number of industrialized countries as well. Singapore has long invested its pension fund resources in private sector assets, as has Malaysia. Chile began restructuring and privatizing its pension system nearly 15 years ago. Argentina, Bolivia, Colombia, and Peru are now following suit. These efforts have been triggered, inter alia, by the need to substantially increase domestic savings, reduce future deficits incurred by large government pension obligations, and curb evasion of payments into the social security system. Major reforms appear to be somewhat easier to put into effect in developing or emerging economies than in more industrialized nations, because the percentage of their populations in the active workforce is high relative to the number of retirees. The transition cost from pay-as-you-go (payg) plans to funded plans is relatively low compared to the cost that would be incurred by most developed countries with larger portions of pension recipients.

The "pension privatization revolution" involves a wide range of techniques: mandatory savings plans, government managed mandatory schemes run like private investment funds, employer-based private systems, and individual retirement accounts. Chile and Argentina have instituted mandatory personal savings plans. Australia and Switzerland use mandatory employer-based pension plans. Denmark and Holland have employer-based plans that are virtually mandatory.

The larger industrialized countries are in particular need of pension reform. For most, the present discounted value of social security pension benefits that have been and are being promised to current retirees, and to men and women now in the workforce, greatly exceeds their conventional debt levels as a percentage of gdp. For some the amount is between 100 percent and 200 percent of gdp. This poses a critical dilemma. Alternative policy courses are to: significantly raise payroll taxes on current and future workers; cut expected pension benefits to large numbers of citizens (through measures such as affluence tests, lowering the indexation escalator, or extending the retirement age); or reduce funding for other government programs to free up funds for social security pensions.

Another, and in most cases, parallel strategy is to fund and raise the investment returns on pension reserves in order to increase the money available to pay future benefits and to decrease the contribution rate needed to support the system. Raising the return on assets from 4 to 8 percent can cut the payroll tax required by nearly two-thirds (see chart).

One major caveat is required before we become too enthusiastic about privatizing social security pensions. Financial markets rise and fall. A sustained bull market in stocks and/or bonds is not an immutable law of economics. Systems that reduce the guaranteed replacement rate of the social security system while allowing a larger portion of the system to be funded and privately managed will be popular when returns are high. But if markets experience a period of sustained weakness, large numbers of retirees will see the "privatized" portions of their anticipated pensions drop substantially--pushing their benefits down to the minimum, far below their expectations. Governments will be subject to intense social and political pressures to appropriate new funds to increase social security payments in such down periods. That is a reality democratic governments must confront as they proceed.

This paper will address principally the investment issues raised by pension reform.

Most countries today are considering whether and how to implement funded, defined-contribution plans. The private sector has led the public sector in this process. In assessing how funded national social security pension funds will be managed, it is useful to examine how private, funded pension plans are managed in the most advanced economies. With this as a guidepost, one can conclude that the process of privatizing and funding national social security plans will likely lead first to investment in short-term fixed income assets, then longer-term domestic bonds, then a broader combination of stock and bonds, and ultimately a still broader diversification into international stocks and bonds.

But for this to happen, domestic capital markets must develop in tandem with pension reform; increasing the pool of investable pension assets can help to support more liquid and dynamic domestic capital markets, but itself is not sufficient to ensure their viability and credibility(which will require legal, regulatory, and institutional reform in many countries. In Asian emerging economies, where stock markets are more developed than bond markets, equity markets will be the first beneficiaries of funded pensions, but pressure will grow from institutional investors to develop deeper and more liquid bond markets. In Latin America, where bond markets are more developed, these will be the first beneficiaries. Diversification internationally will be necessary over time in order to avoid excessive concentration of risk in the home country by individual pension funds. Even in countries with large current account deficits, that is net capital inflows, individual pension funds will likely need to diversify internationally to some degree because of the risk of placing all of their assets in their home market and therefore running the risk that political upheavals, poor economic policy, or other factors could disrupt their markets, leaving their pensioners with a low or negative return. Emphasis will be on investment in foreign markets where performance is not closely correlated with the domestic market.

Industrialized Countries

There are many similarities in the demographics/ pension circumstances of the major industrialized countries (Chart). All have aging workforces and most have large unfunded liabilities. But there are important differences. The type and quality of management of private pension funds varies greatly (see chart). And government social security plans also vary greatly in character and size of liabilities. Using oecd assumptions of 1.5 percent productivity growth, U.S., U.K. and Canadian social security pension expenditures peak at between 5(8 percent of gdp compared with expenditures of between 15(20 percent of gdp for Japan, Germany, France, and Italy. This variance reflects differences in both the size of elderly dependency ratios and the generosity of benefits. And despite higher contribution rates in the last group of countries, the gap between contributions from currently employed workers and payments to pensioners is much greater for them than for the first three. The first three above also have well funded private pension schemes.

The above increases, plus larger health care payments also related to aging populations-and the growing cost of debt servicing-will lead to enormous increases in net public debt levels in the early decades of the next century. According to oecd projections, U.S. net public debt will rise from around 40 percent of gdp in 2010 to 105(120 percent in 2030 (depending on interest rate assumptions). In Japan net debt will grow from 25 percent in 2000 to 255(315 percent in 2030. In Germany, the growth will be from 45 percent to 80(105 percent, and in France from 40 percent to 75(105 percent.

The global impact of these prospective increases in unfunded government pension liabilities is enormous. Growing deficits will absorb a huge amount of national savings-thus diverting massive amounts of funds from other productive activities and placing great upward pressure on interest rates.

For the United States, which continues to borrow large sums from abroad to finance its current account and its budget deficits, these considerations will in time severely limit its ability to do either, unless of course emerging nations with higher rates of savings are willing and able to finance such deficits (although, as shown below for China, demographic problems later on will place constraints on them as well). Of course other industrialized countries also will face external financing constraints similar to those of the United States. Overall, such considerations will place the economic and social systems of all major economies in peril, triggering growing intergenerational tensions.

There are strong arguments for the major industrialized countries individually to take bold corrective action, but because the pension crisis is an international problem there are similarly strong arguments for them to proceed in parallel. Doing so would, at the margin, help national leaders to mobilize domestic public opinion for adjustments at home. Each could argue that citizens of all industrialized countries were making difficult adjustments. This would by no means overcome domestic resistance to dramatic or painful changes but on balance could be helpful. The subject would be appropriate for priority consideration by the Group of Seven summits; leaders could share experiences in reforming their systems and in how to present needed changes to their electorates.

I now turn to three of the most important countries that will need to reform their pension plans in coming years and the implications for capital flows of the measures each will likely take-toward defined contributions and greater international asset allocation (see chart).

Japan

Japan is the most rapidly aging nation in the world today. Fully one-fourth of the population is expected to be over the age of 65 by the year 2015. How Japan copes with this "demographic tsunami" will profoundly affect the welfare of all Japanese, corporate balance sheets, and domestic/global markets for stocks and bonds.

Currently Japan's pension assets are dominated by public pension funds. These pension funds amount to 142 trillion yen, or 73 percent of the total, while private corporate pension funds amount to 51 trillion yen, or 27 percent of the total. This compares with public/private pension ratios of 32:68 in the United States, 15:85 in the U.K., and 39:61 in continental Europe (although the ratios are higher in Germany and France(Germany is an anomaly since private pensions are mainly not pre-funded in an asset sense). Public pensions in Japan are divided into the National Pension, which provides a basic fixed benefit, and employee pensions (epi), which provide an earnings-linked benefit. Both schemes are regulated by the Ministry of Health and Welfare.

Combined, these are analogous to the Social Security system in the United States.

Private corporate pension schemes are of three types: employee pension funds (funded by the employees and employers), tax-qualified pensions (funded by sponsoring firms), and non-tax qualified pensions (which are not funded in advance and which are considered part of the plan sponsor's overall business).

One of the biggest differences between Japan's system and that of other countries is that corporate pension fund assets are held in the custody of trust banks and life insurance firms and do not get market-based returns. Aside from this, Japan's pensions are similar to those of several European countries in terms of high dependency on public pensions, a relatively high social security replacement ratio, numerous asset allocation restrictions, and minimal disclosure. At the other end are the United States and U.K.; they have a higher dependence on private pension funding, lower replacement ratios, fewer portfolio restrictions and fairly stringent pension funding and disclosure requirements. And in the United States, private defined contribution plans have become fairly widespread, comprising roughly one-third of total pensions.

Japan needs to make major changes in its pension programs. By 2015 nearly one-quarter of the population will be over 65, compared to around 17 percent for Western Europe and 15 percent for the United States. A shrinking work force, growing aged population and poor investment returns will put intense pressure on pension funding as the number of beneficiaries gradually approaches the number of contributors. The ratio of active contributors to beneficiaries in the pension system is expected to drop to 2 by 2030, compared to 4.4 currently, well below the 2.8 expected for the United States.

So far, changes have included raising the retirement age from 60 to 65 by 2013 and increasing the contribution rate for employee pension insurance from 14 percent to 16.5 percent of monthly salaries, with plans to raise it to 29.6 percent by 2025-nearly double the current rate. The government of Japan believes these measures will be sufficient to prevent a deterioration in government pension finances. But government expectations depend on optimistic assumptions, which include anticipating that investment returns of 5.5 percent can be realized. Under current conditions this is unlikely. The oecd believes that rising Japanese pension expenditures will result in a budget deficit to gdp ratio of almost 20 percent by 2030-higher even than the nearly 13 percent projected for Italy.

Additional steps have recently been instituted to reform the system, including a shift in accounting for private employee pension funds from book value to market value, allowing one-third of such funds to be managed by independent investment advisors, including non-Japanese firms, allowing investment advisory firms to manage a portion of assets of the Pension Welfare Service Corporation (Nenpuku) and allowing investment advisory firms to manage up to one-third of corporate pension funds that have been in existence for more than three years (instead of eight as had been the case). Further liberalization is inevitable, and with it is likely to come a shift toward equities, both domestic and foreign.

Japan has a window of opportunity to implement necessary changes in the structure of its pension system. Its pension problems are not primarily associated with a lack of funds; rather it is the lack of efficient allocation of these assets that must be addressed. Reforms announced this week should enable Japanese fund managers the flexibility to produce higher returns. They will likely to lead to increased investments in equities and foreign assets.

to drop to 2 by 2030, compared to 4.4 currently, well be

Germany

Germany, like Japan, will experience a sharp aging of its population in coming years. German population is expected to drop from around 80 million today to below 70 million by 2040. The ratio of people over 60 to those in the 20 to 59 age group is expected to rise from roughly 35 percent to 70 percent in 2040. Contributions to pension insurance are forecast to rise from just under 20 percent of gross wages currently to nearly 28 percent by 2040 under scenarios of moderate immigration.

A number of changes were made in 1992, setting pensions at about 70 percent of net wages and adjusting them in line with net wage growth instead of gross wage growth as before. And the government is raising the retirement age to 65. More recently the government has changed the entry age into retirement because of unemployment from 60 to 63.

Additional steps are likely to include: a further reduction in the government pension replacement rate, a further increase in the retirement age, and lower growth of net pensions. This shift will place additional emphasis on company pensions and individual savings. Deutsche Bank was, perhaps, a trendsetter among German companies in establishing a formal pension fund for its employees that will invest about one billion deutsche marks in securities over the next two years, although IBM Germany's pension has been funded for many years.

Germany's pension system is evolving from heavy reliance on public benefits provided on a PAYG basis to a system with greater emphasis on individual and company pension funds; this will also require a change in the taxation of contributions and benefits-in particular eliminating the strong bias against private savings for retirement and the yields of these savings. Also, in Germany corporations can take tax deductions for pension contributions without actually funding the pensions; whereas in the United States, U.K., Canada, and Japan a corporation must fund to receive a deduction-an anomaly which is likely to be subject to scrutiny in the future.

One consequence of changes in Germany in coming years is likely to be an increase in the share and amount of investment going into equities. For a number of historical reasons German investors have a strong aversion to equities. As the attached chart indicates, German pension assets are largely concentrated in bonds and invested less in equities than those of any other major economy. So growth in equity investments is likely as pension funds increase; it is also likely that investment in international assets will grow as pressures for diversification emerge.

China

China has become a major participant in international financial markets and a large buyer of American Treasury Bonds. It is the largest economy in Asia outside of Japan and its demographic circumstances will mirror those of Japan-with a very considerable lag. The dependency ratio of pensioners relative to active workers is low now, because of a relatively young population But the country's policy of one child per family means that the dependency rate will rise dramatically as, in effect, two retired parents must be supported by one working child. According to World Bank estimates, the dependency ratio will double from roughly 20 percent today to over 40 percent in 2030 to 55 percent in 2050 to over 70 percent in 2100, based on current projections.

Most workers covered by China's currently very generous pay-as-you-go pension plans are in state-owned enterprises and relatively old cooperatives( which, all told, amount to about 25 percent of the working population. The high replacement rate in the SOEs has become a problem as they have in many cases lost competitiveness, and must pay high pensions (and often high payroll taxes) and maintain large and often inefficient workforces. The average replacement rate is over 80 percent, as opposed to the 40 to 60 percent of most other countries. And the duration of retirement is long because retirement ages are low relative to rising life expectancy. Most urban workers in non-state corporations and virtually all agricultural workers are not covered; if the extended family system breaks down, the latter will become vulnerable.

According to the World Bank, China is interested in moving to a partially funded system to avoid the high future tax rates of a payg system, as well as to build up a stock of assets in retirement savings funds; in the process it can use this system to develop domestic financial markets and finance long-term investments. So far, developing a funded system has proved an elusive goal because contributions need to be used to pay current pension obligations.

The World Bank has recommended that China shift to a "multi-pillar" system that includes: 1) a funded, privately managed defined contribution pillar, and 2) a payg publicly managed defined benefit pillar to provide a social safety net. The funded pillar can serve to sustain China's high rate of savings and support financial market development. That will be important to break out of a pension system now subject to significant local government control, with funds largely invested in central government bonds and few alternative financial opportunities. It will become increasingly necessary to ensure that funds are channeled to investment that produces a high rate of return both to reduce future contribution rates and to provide a reasonable replacement rate. This will likely mean a shift to equities over time. The prospect of a large shift to international investment will be limited as long as domestic returns continue to be high-but some international diversification on a fund-by-fund basis is likely.