You may be interested in my comments on the proposed Pension Accounting and Financial Reporting by Employers standard being proposed by the Government Accounting Standards Board.
Director of Research and Technical Activities
Government Accounting Standards Board
Project No. 34
The Plain Language Supplement for Pension Accounting and Financial Reporting by Employers makes a strong case for treating pension benefits owed to employees of state and local governments as liabilities of their employers. But I do not think the section on investment earnings is nearly as strong as it should be, and here is where the tragic surprises are most likely to occur.
The conventional accounting method, which the proposal supports with the corridor approach, is to treat future earnings from portfolio investments as if they could be measured by annual rates. But this over-simplified method does not track with how stocks and some other investments will provide the cash that must be raised to pay retirement benefits, so it can be seriously misleading, particularly if asset prices decline.
Most securities in retirement portfolios fall into one of two categories, those in which some party has an obligation to provide specified amounts of cash to the portfolio at specified times, and those which are in an inventory of things that were bought to be sold at a profit. To keep this discussion simple, I will refer to the two categories as bonds and stocks.
The annual rate method may be well suited for bonds that a portfolio holds in order to receive regular payments of interest that will be used for pension benefit payments. If ladders of bonds have been assembled whose maturities are intended to coincide with pension benefit payments, the annual rate approach may not show how well the bond maturities and pension obligations match, or how well the portfolio can be expected to serve younger retirees after the bonds have been retired or sold.
The annual rate method is not suitable for stock portfolios because it hides the realities of how stock investments produce cash. Except for the few that pay significant dividends, most stocks are commodities that are bought and held in inventory for eventual sale at a profit. When I learned accounting, a basic principle was to value business inventories at cost or market, whichever was lower, and a cardinal rule was that one did not accrue profit. To mark up inventory values before the goods were sold was to accrue profits that had not, and might never be made.
We see the consequences of ignoring that principle and rule in real estate markets today. We also see them in retirement plans. At about 1100 as I write this, the S&P 500 is 25 percent below where it was at the end of 1999. Portfolios that were marked to market at that time, suffered major losses that changed the retirement question from “when” to “if” for many baby boomers. One of the worst things that could happen in the near future would be for stock prices to return to their former highs and restore confidence in stock-based retirement plans.
Stock-based retirement plans, including company defined benefit, 401(k), 403(g), and state and local government pension plans, have at least eight fundamental flaws.
They are built on a stocks-for-retirement cycle. Most baby boomers are in the front, or buying half of the cycle, and to receive retirement incomes, they will have to shift to selling their stocks for substantial gains in the back half. This makes the boomers’ cycle a national Ponzi scheme because returns to early investors (the boomers) must come from money paid in by later investors (younger workers), not as dividends paid from company earnings.
When boomers gradually shift from buying stocks to selling them, the primary domestic buyers will have to be the same younger workers that some believe will not be able to sustain Social Security in its present form. Stock-based plans and Social Security are joined at the hip by the same demographics—if stock-based plans can work, there will be no Social Security problem, but if Social Security can’t work, neither can stock-based plans. Despite urgings to boomers to save and buy more stocks, the future of their plans will be determined as much if not more by the saving and stock-buying habits of younger workers and their retirement plans as by the boomers’ own actions.
There has been a symbiosis between corporate insiders and boomers’ retirement plans. The plans wanted the insiders’ stocks and the insiders wanted the boomers’ money. As the boomers’ plans shift from buying to selling, the relationship will end, and they will have to compete with insiders who will still be selling. This will add more downward pressure to stock prices in what may become a sustained bear market.
Boomers are told to plan to stretch their stock sales over many years, but if there is a serious bear market, some of them may be forced to liquidate or decide to get out quickly and save what they can. This, of course, would feed the bear.
Boomers’ retirement plans that will have to sell stock suffer from the fallacy of composition; while some might be able to build and store future purchasing power individually, all of them cannot do it collectively. Retirement income cannot be stored for a whole generation. It is a flow that can only come from other flows like employee and company earnings or taxes, which is why there appears to be a Social Security problem.
If there were an accepted due diligence analysis or feasibility study that explains how the boomers’ stocks-for-retirement cycle can be expected to work, the financial services industry would quote it like a mantra. But there is no such explanation, so when asked, Wall Street changes the subject.
Stock-based plans suffer from the same fatal mistake of anticipating ever higher asset prices that led to the housing bubble. There is no accepted explanation of how stock prices can increase faster than the economy grows for several generations, but this must happen for younger workers to pay adequate prices for the boomers’ stocks and then sell them at a profit to pay for their own retirements.
In a 2004 paper titled “Demography and the Long-Run Predictability of the Stock Market,” John G eanakoplos of Yale and two associates on the West Coast explained that there has been a close relationship between stock prices as represented by price-earnings ratios and the ratio of young and old adults in the population. They predict a long bear market when boomers’ plans switch to selling.
The termination of thousands of company pension plans and the sorry shape of many state and local plans are evidence of these flaws. Any one of the flaws should be enough to make economists, corporate and government officials, and Wall Street question boomers’ plans before they all fail. But instead of answering feasibility questions, they avoid them. Adopting the corridor approach would sustain the fiction that stock-based retirement plans have a chance of doing what has been claimed for them.
Retirement plan managers expect company executives to understand their businesses which includes knowing the sources of their products, the markets into which the products will be sold, and how they expect to earn profits. By this simple standard, few retirement plan managers understand their own businesses, and so far, accounting standards have not exposed this failure.
It is revealing to look at all retirement plans as a single national business. It is not very complex. My analyses show that at the end of 1981, they owned 29 percent of the domestically-owned stocks traded on U.S. markets. By the end of 2008, they owned 64 percent.
There are strong reasons to believe that retirement plan purchases during the buying half of the boomers’ stocks-for-retirement cycle contributed to the rise in stock prices, price earnings ratios, and price dividend ratios until they peaked after 1999. Because the primary sellers of the stocks were the households of corporate insiders, the purchases also increased the wealth gap by transferring middle class retirement savings and pension plan funds to the richest people in the country. That business model is not likely to be sustainable, either economically or politically, and it may be failing even before most boomers reach retirement age.
The most realistic way to forecast investment earnings is to project the total stock sales and that can be expected to pay for boomers’ retirements, project the demand for those stocks, and anticipate the prices at which they will be sold. This cannot be done with precision, of course, so the projections should be made as ranges, not specific numbers. Some of the ranges will be negative. Unless someone can explain how the stocks-for-retirement cycle can produce profits for decades that are substantially greater than the national growth rate, the ranges will show the risks that not only is it a mistake to expected stocks to help state and local governments meet their pension plan obligations, they may increase burdens on the governments. And the ranges will show the futility of trying to project future earnings for specific pension plan stock portfolios on basis of past performance that does not include a period of massive liquidations.
This will be particularly important for state and local government pension plans because as you state, most of them have infinite lives. The rate of return method of accounting hides the fact that as they operate today, they will have to simultaneously sell stocks to raise money to pay retired boomers while they buy stocks to build portfolios for their younger workers. There will be many wash-transactions that do not produce more cash. If the plans receive high prices to pay benefits to boomers, they will pay high prices that will require even more price inflation to serve the younger workers.
Thinking this through on a national basis will demonstrate the futility of treating what is coming as if it could be measured with annual earnings rates for individual units of government.
Thank you for this opportunity to comment, and I will be happy to discuss this with you or your staff.
1115 Glenmoor Drive
Harrisonburg , VA 22801