Investment returns must improve. The security of our Health Care Stabilization Fund and our pension benefits depend on growing our portfolio. Our pension fund went from about $80 billion to about $55 billion in less than 18 months. Even if our fund is able to recover, the 8.4% projected long-term returns are too modest to justify 30+% volatility.
A large portfolio always needs an appropriate asset allocation model and markets with high-volume and high-liquidity. Even at $55 billion the STRS fund is a large portfolio.
Appropriate Asset Allocation -- The asset allocation model is usually considered to be the primary determinant of the investment performance of a large portfolio. A fund with payout responsibilities, such as pension and health care benefits, needs some of its income in cash. Portfolio strategists divide the investments of the portfolio into "asset classes" such as "equities" (company stock), "fixed income"(usually bonds and CDs), "real estate", "cash" and others. The investments in each asset class tend to have similar investment performance. The strategist attempts to allocate investment funds among the asset classes so that the returns on the total portfolio will be less volatile and slightly higher than if all the assets were invested in one class. The perfect strategist could assign the asset allocations to adjust the volatility and return to the target levels. Often high short-term volatility means higher long-term returns. This is one origin of the "high risk -- high reward" cliché. The strategist will ideally try to find asset classes with high long-term returns and a perfect negative correlation to each other. So when Class A is suffering losses Class B will be more than making up for those losses. Since there are no perfect positive or negative correlations, other than taking short positions, the strategist analyzes correlations between many asset classes to move toward this goal. Asset allocations are often displayed as pie charts in reports.
The STRS asset allocation model is outdated and imprudent even though it is similar to the policies of other troubled pension funds. The asset allocations STRS has been using cause performance to follow the US stock market and the world value of the US dollar too closely. I suspect that this is true because our board members and investment advisers, like the board members and advisers of the other funds, were trained in investment philosophies that were proven valid and became popular when the stock markets of the United States were the principal investment venue for the world. At that time, the US stock and bond markets offered the most liquidity of any markets in the world, the value of currencies were somewhat fixed and did not change value against each other every second, there were no 24/7 investment markets, and very few traded or invested outside their own countries unless they invested in the United States. When some of the financial advisory firms were founded there were people who would rather have a paper US dollar than the gold or silver it (then) represented.
In the modern world, the stock markets of the United States are instantly affected by global financial and political developments of all kinds. Frequently, when the value of the US dollar declines significantly, there will be idle ramblings in the trading or financial press about what could replace the US dollar as the world's reserve currency. There are no current serious challengers but the entire concept was unthinkable 50 years ago.
Our managers must have a new asset allocation model that allows them to compensate for currency related and trade-related fluctuations with origins outside the United States. These are not taken into sufficient account by our current asset allocation. It doesn't help to make 5% on your investments if the purchasing-power of the currency in which you hold those investments declines 25%. This is approximately what happened to investors who held "safe" U.S. Treasury bills between 2000 and 2005.
Better-performing areas of large portfolios of successful pension funds, foundations, and large investors around the world continuously update their asset allocation models to reflect changes in global finance. STRS has ignored important developments and is too heavily invested in dollar-denominated investments, particularly US stocks and fixed income instruments. Five of our six asset classes are almost exclusively US dollar denominated. I suspect that even the "International Equities" class contains ADR shares that are denominated in US dollars. As much as Five-Sixths of our eggs are in one basket.
The range of asset classes used by STRS must be increased. STRS lacks significant exposure to entire uncorrelated asset classes that would cut our overall portfolio risk and increase returns. When the US stock market is trending down there are other asset classes and other markets that trend up. The board must encourage our asset allocation strategists to consider their appropriate use.
Market Volume and Liquidity -- The fund needs to trade or invest in markets that are large enough so the investments the fund makes are not too big for the market to absorb. There must be a sufficient number of both buyers and sellers in the market so that purchases and sales are being made often enough for the portfolio to be evaluated. An analogy would be to go to a garage sale with $100,000 and expect to be able to spend it all without influencing prices -- only one buyer, one seller, and lots more money than goods available for purchase. However, you would have better volume and liquidity at a well-attended automobile or real estate auction and would be able to spend all your money without driving prices up.
The problem of "commonly trained advisers" mentioned above also affects the liquidity that is available to our fund. If all the advisers want to make similar investments then you have lots of funds trying to get into or out of the same chosen market at the same time. It's like trying to use 25 row boats in a swimming pool. There's not enough liquid. Improving our asset allocation could turn this problem into an opportunity.
It's interesting to think that involvement in global equities will be sufficient to improve our ability to diversify. The diversification available to a fund of our size may not live up to this expectation. The top 10 companies listed on US and Japanese exchanges account for between 20% and 35% of the market capitalization traded. The largest 10 stocks traded on eight other major stock exchanges represent between 45% and 70% of the capitalization of those markets. The largest companies also trade in all markets. While global diversification is necessary, we may not be increasing our opportunities significantly by simply expanding into global equities.
Specific recommendations:
Eliminate the "international" asset class. In an increasingly global economy with transnational and perhaps extra-national corporations, the existence of an "international" asset class is quickly becoming an illusion. Analysis of investments on a global basis (below) also makes this class unnecessary.
Eliminate the "alternative investments" asset class. Because of increases in exchange-traded market opportunities and other global changes, I believe that the theoretically enhanced performance of “alternative investments” can be achieved with more liquidity and less risk within other asset classes. The nature of the investments usually contained within this class is sufficiently broad to make correlations with other classes uncertain and risk assessment questionable. STRS's percentage allocation to this asset class is also so small that staff time used in analyzing these investments would likely provide more benefit if used elsewhere.
Analyze all asset classes on a global basis. There are enough quoted market indices that major asset classes can be tracked and compared on a global basis using the Stable Currency Benchmark or a suitable substitute. The Stable Currency Benchmark is a basket of currencies specifically designed to monitor the global purchasing-power value of any asset at any point in time. This allows the accurate comparison of performance between investments valued in different currencies and would automatically tend to reduce our exposure to US dollar denominated markets at inappropriate times.
Introduce opportunities for perfect negative correlation within each class by using the absolute value of the percentages in the asset allocation targets. If the asset allocation requires 10% exposure to an asset class or subclass, strategists would charge investment managers with moving toward either a +10% or -10% exposure to that asset class using whatever financial instruments might be necessary. There is enough global liquidity in appropriate instruments to implement this strategy for the major asset classes.
Add "foreign exchange" as a risk adjustment tool and perhaps an asset class. At $3 trillion per day, the electronic foreign exchange market trades roughly 10 times the value of all the stock exchanges in the world combined. Transaction costs are small compared to other investments and the markets are completely transparent. As a risk adjustment tool, the foreign exchange market represents a pool of liquidity that will allow our strategists to quickly lay off risk from other parts of the portfolio. It can be used to assist managers working in the other asset classes in implementing the "absolute value" policy (immediately above). As an asset class foreign-exchange offers long trends and the liquidity for managers to move between long and short positions safely without deforming the market.
After investment returns have been smoothed and improved, the Health-Care Stabilization Fund adequately funded, and the pension funding period reduced, retirees’ purchasing power could be maintained with retirement matrix improvements, improvements to the COLA formula, or even reinstatement of the 13th check. 
