Remarks by Governor Susan Schmidt Bies
To Financial Executives International Baltimore Chapter, Baltimore, Maryland
January 18, 2005
The Economy and Challenges in Retirement Savings
I am very pleased to join you for this meeting of The Baltimore Chapter of Financial Executives International. Today I want to begin with a brief assessment of the economic outlook before discussing financial conditions of households and businesses in more detail. Then, I will turn to some important trends in retirement savings, including the responsibilities that both businesses and households have in planning for the financial security of workers later in life.
I also need to add that I am expressing my own opinions, which are not necessarily those of my colleagues on the Board of Governors or on the Federal Open Market Committee.
The Economic Outlook
On the inflation front, broad measures of consumer inflation have risen somewhat faster than they did in the year-earlier period, boosted by higher energy prices. Focusing on the core price index, which excludes food and energy, it has been a bit higher than it was the year before. This reflects changes in the pattern of prices of goods (as opposed to services) purchased by consumers, as prices of goods that had been falling in 2003 began to stabilize and rise slightly in 2004. These effects should taper off and I expect that core inflation will remain in its current range. Moreover, surveys indicate that inflation expectations over the longer-term appear to have remained well-anchored. I believe that, with underlying inflation expected to moderate, the Federal Reserve can continue to remove its policy accommodation at a measured pace, consistent with its commitment to maintain price stability as a necessary condition for maximum sustainable economic growth.
Household Financial Conditions
It is true that households have taken on quite a bit of debt over the past several years. According to the latest available data, total household debt grew at an annual rate of about 10 percent between the end of 1999 and the third quarter of 2004; in comparison, after-tax household income increased at a rate of about 5 percent over this period. This rapid growth in household debt largely reflects a surge in mortgage borrowing, which has been fueled by historically low mortgage interest rates and strong growth in house prices.
Indeed, many homeowners have taken advantage of low interest rates to refinance their mortgages, some having done so several times over the past couple of years. Survey data suggest that homeowners took out cash in more than one-half of these "refis," often to pay down loans having higher interest rates. On net, the resulting drop in the average interest rate on household borrowings, combined with the lengthening maturity of their total debt, has damped the monthly payments made by homeowners on their growing stock of outstanding debt.
The Federal Reserve publishes two data series that quantify the burden of household obligations. The first series, the debt-service ratio, measures the required payments on mortgage and consumer debt as a share of after-tax personal income. The second series, the financial-obligations ratio, is a broader version of the debt-service ratio that includes required household payments on rent, auto leases, homeowners insurance, and property taxes. Both ratios rose during the 1990s, and both reached a peak in late 2002. Since then, however, the debt-service ratio has been stable and the financial obligations ratio has receded a bit, an indication that households, in the aggregate, have been keeping an eye on their financial commitments. Consistent with these patterns, delinquency rates for a wide range of household loans either have drifted down over the past year or held about steady at levels below recent highs.
The low interest rates of the past few years, however, will give way as the economy continues to expand, and we have already seen an uptick in mortgage rates and on some other consumer loans during this past year. To be sure, some households will be pressured by the higher rates, but I believe that concerns about their effect on repayment burdens can be overstated. First, most household debt--mortgage and consumer debt combined--carries a fixed interest rate, which slows the adjustment of interest costs to rising rates. Second, although interest rates on some variable-rate loans will rise quickly, the adjustment for a large number of variable-rate loans could occur rather slowly. For example, many adjustable-rate mortgages start off with a fixed rate for several years, providing households with some protection from rising rates.
Another concern is that house prices will reverse and erase a considerable amount of home equity built up in recent years. Recent gains in house prices have been notable: the average house price rose 13 percent in the year through the third quarter of 2004, and cumulative gains since 1997 now top 60 percent.1 Despite a rise in mortgage debt, the current loan-to-value ratio for outstanding mortgages is estimated to be around 45 percent, roughly the level that has prevailed since the mid-1990s. It is true that some households have considerably less equity in their homes, and these households tend to have lower income and fewer other financial assets to cushion shocks. Based on the 2001 Survey of Consumer Finances (SCF), a small share, 7 percent, of households had a loan-to-value ratio of 90 percent or more. Unfortunately, we cannot characterize the current share as accurately, at least not until the 2004 survey becomes available early next year, but it is unlikely that the share has risen by a lot. While new originations of mortgages with high loan-to-value ratios in recent years would push this share up, the substantial house price appreciation in that same period likely improved the financial positions of the households with high loan-to-value ratios in 2001.
This relatively upbeat assessment of household credit quality seems to be shared by lenders and by investors in securities backed by consumer debt. According to the Federal Reserve's survey of senior loan officers, banks were in a relatively neutral stance, neither tightening nor easing on net, with respect to lending standards for consumer loans or mortgage loans through most of last year. Moreover, credit spreads on securities backed by auto loans and credit card receivables have narrowed in recent months.
Some analysts have also expressed concerns about the decline in the personal saving rate. Aggregate personal saving, measured by the Bureau of Economic Analysis, averaged about 1 percent of disposable income during the first three quarters of 2004, more than 6 percentage points lower than the average that has prevailed since the early 1960s. The saving rate, measured by the Board's flow of funds accounts is higher, at 5 percent of disposable income, though it, too, is significantly lower than its average level in the past. Analysis by Board staff using data from the SCF indicate that households in the top income quintile can account for nearly all of the decline in the aggregate saving rate since 1989. Given that these higher-income households have more financial resources to weather shocks, the significant decline in savings is less troublesome than if it had occurred in the lower part of the income distribution.
This points out two different perspectives on household financial health. While analysts usually focus on the savings rate as a share of current income and funds flow, some argue that a more relevant measure of saving adequacy is the change in net worth. And in this regard, the picture of household saving looks more favorable than suggested by the saving rate. The ratio of net worth-to-disposable income has come down from its peak in 2000, but remains at a high level relative to the past few decades, because capital appreciation on household assets, such as equities and real estate, has considerably outpaced income gains. This is a passive perspective on savings, though, where households rely on the markets to raise the value of their assets over time. But to create these assets, households need to consistently set aside some of their current earnings to invest for their future needs. While the experience of the past three years of exceptionally low interest rates and lower expected stock returns encouraged a rational consumer to spend and not save, as the markets return to more long-term trends, we should see consumers moderate their behavior as well.
Financial Conditions of Businesses
The improvement in corporate balance sheets in the past few years has been substantial. Firms have taken advantage of low long-term interest rates to refinance high-cost debt. Businesses have also improved their balance sheet liquidity by substituting long-term debt for short-maturity debt and by building up their cash positions. In addition, many firms--especially in the most troubled industries--have retired debt through equity offerings and asset sales, and others have used their mounting profits to retire debt. As a result, nonfinancial corporate debt grew at an annual rate of less than 2-1/2 percent in the past few years, its slowest pace since the early 1990s.
These repairs to balance sheets have also reduced the exposure of many firms to rising interest rates, especially in the near term. In particular, the replacement of short-term debt by long-term bonds means that less debt will have to be rolled over in the near term at higher rates. In addition, because much of the long-term debt has a fixed rate, interest payments typically are unaffected over the life of the bond. Moreover, research by the Board staff suggests that firms that rely more on floating-rate debt, and for that reason might be more vulnerable to rising rates, have in recent years tended to use derivatives to hedge some of their exposure to interest rate risk. Thus, for many firms, the effect of rising interest rates will be mitigated and spread out over time.
Also, as we learned from the episode of policy tightening in 1994, rising interest rates have little detrimental effect on the financial health of the corporate sector when the rate increases occur in the context of an expanding economy. Indeed, corporate credit quality improved on balance after 1994 with the pickup in economic activity and corporate profits.
Some of the improvement in financial conditions among businesses is due to significant belt-tightening by many firms. Over the past few years, the drive to cut costs and boost efficiency has generated rapid productivity gains. Fuller utilization of the capabilities of capital already in place, ongoing improvements in inventory management, and streamlined production processes requiring fewer workers, to name but a few examples of efficiency enhancements, have boosted corporate profitability.
The pickup in revenue growth since mid-2002, combined with outsized productivity gains, has produced a dramatic recovery in overall corporate profitability. The profits of nonfinancial corporations as a share of sector output continued to climb and reached almost 11 percent in the third quarter of last year. This share lies above its long-run average over the past few decades and well above the cyclical trough of 7 percent in 2001. To be sure, the profit share likely will slip a bit from its high level as the expansion gains steam and businesses hire new workers more aggressively. But some decline in the profit share is to be expected and will not, in my view, significantly impair the financial health of companies. This favorable view is reflected in risk spreads on corporate bonds, which have dropped dramatically from their historic highs in the fall of 2002. And firms that have turned to capital markets for financing have found them to be accommodative.
A remaining financial hurdle for some companies is the underfunding of defined benefit retirement plans, though this burden is notably less than it was two years ago. At the end of 2002, the majority of S&P 500 plans were underfunded, with a net shortfall that had grown to more than $200 billion, as stock market losses from 2000-02 eroded the value of assets and declining interest rates raised the current value of plan liabilities. Since then, companies have made large cash contributions to shore up these plans and equity prices have risen, reducing the aggregate underfunding for S&P 500 companies to around $125 billion at the end of 2004. In addition, plan sponsors received some temporary relief from federal legislation passed early last year that allowed firms to use a discount rate for their liabilities based on corporate bond yields rather than one based on Treasury yields. The replacement effectively reduced the stated value of liabilities and thus estimates of underfunding and required cash payments for tax purposes.
Nonetheless, important longer-term issues regarding defined benefit plans remain. In decades past, workers might spend most or all of their careers at a single firm and might receive generous preset benefits from traditional defined benefit plans, with a guaranteed benefit to be provided by the Pension Benefit Guaranty Corporation (PBGC), the government entity that insures defined benefit plans, if the firm were to go bankrupt. As workers began to change jobs more frequently, however, they increased their demand for defined contribution retirement plans because of their portability and the more-even pattern of benefit accruals over a worker's career. Also, many companies were attracted to these plans as a way to limit their future funding obligations for traditional defined benefit retirement plans. As a result, financially solvent older companies have been terminating their defined benefit plans and only a rare few companies have been starting new defined benefit plans. This trend weakens the PBGC because it is left increasingly with pension plan sponsors with relatively weak financial positions.
Currently, the PBGC is facing a potential financial crisis. Its funding status deteriorated in 2004, for the fourth consecutive year, with the value of its assets now falling short of its liabilities by nearly $24 billion. Most of the deterioration is due to the large number of firms with underfunded pension plans that declared bankruptcy in recent years. However, the legislation that offered relief to plan sponsors also contributed to the worsening position because it reduced not only the cash contributions made to plans but also the premiums to be paid to the PBGC. The PBGC has fundamental structural problems as well, which include the inability to raise premiums or to charge premiums based on the credit quality of the plan sponsor, and the lack of authority to mandate adequate contributions to defined benefit plans. These flaws, if left unaddressed, combined with the shift toward defined contribution plans, raise serious questions about the retirement security of workers currently under defined benefit plans.
A particularly popular type of defined contribution plan is the 401(k), which lets workers make pre-tax contributions to retirement accounts through payroll deduction. Twenty-five years ago, such plans did not exist. Today, they cover more than 40 million workers, take in $150 billion in annual contributions, and hold assets of about $2 trillion. And with a decline in the number of workers being covered by defined benefit plans over this period, the growth of 401(k) plans has put greater responsibility on the part of individuals to actively manage their retirement wealth. For example, workers must decide whether to participate in the plan, how much to contribute every pay period and in what type of assets, and when to rebalance their asset mix. In addition, they need to decide what to do with balances when changing jobs.
Economic theory provides the basis for making these types of choices, but the average employee may not have developed the skills or the interest to formally evaluate the alternatives. And, in fact, recent research has found some troubling patterns: Many workers do not participate, contribute only a small portion of their wages if they do participate, and make questionable investment and distribution choices.
Let me take a couple of minutes to cite some telling facts about individual savings in 401(k) plans, suggesting that workers may not be giving adequate attention to their savings in the retirement plans sponsored by their employer. First, despite the tax advantages of 401(k) contributions, one-quarter of workers eligible for 401(k) plans do not participate at all, even if the employer would match a portion of their own contributions.2 These workers are effectively giving up a pay raise. And among those that contribute, many save just a little. In a survey last year, one-quarter of firms reported that their rank-and-file 401(k) participants saved an average of less than 4 percent of pay.3
Another concern relates to the way employees manage their 401(k) plans. Some participants simply invest contributions equally across the investment options or according to plan defaults, which in many cases is a low-risk, low-return money market fund.4 And, as has been publicized widely in recent years, many 401(k) participants invest heavily in employer stock. Among companies that offer company stock as an investment option, more than one-quarter of 401(k) balances are in company stock.5 This high concentration cannot be attributed entirely to an employer match that is required to be held in company stock. Instead, employees appear to voluntarily purchase abundant amounts of company stock, despite the obvious risk of linking their current income and retirement wealth to the financial health of their employer.
These patterns are troubling because they raise doubts about the financial security of workers in later life. Fortunately, new research in the discipline of behavioral finance provides some important insights into the behavior of 401(k) participants and suggests some promising changes that can lead workers to make savings choices that will leave them better prepared for retirement. Contrary to predictions of traditional finance theory, the way the retirement-plan options are framed affects the choices made by participants. As compelling evidence that framing matters, researchers have found that "opt-out" plans--those that automatically enroll workers unless they actively choose not to enroll--have substantially higher participation rates.6 Moreover, when defaults are designated, many workers tend to enroll using the default contribution rates and investment options and to leave these in place for many years after enrollment, even though the default may be set too low to allow for the accumulation of sufficient retirement assets.7 Moreover, the default investment choice is often a low risk asset that, while safe, does not allow for sufficient expected returns to build up retirement wealth.
If workers are influenced by how choices are offered, then employers can make changes to the plans to help participants make better decisions. For example, employers might set default contribution rates to rise as workers receive pay raises, or set the default investment option to a diversified portfolio that adjusts as the worker ages.8 In addition, employees have increasingly expressed interest in employer-provided financial education, and firms are responding. In this regard, the Congress passed legislation in December 2001 that removes obstacles for employers to hire a third-party to provide financial advice. Previously, firms had been reluctant to provide investment advice for fear of being held liable should such advice lead to losses, even if an adviser was considered to be competent. Reportedly, more firms now provide access to a third-party who will, for a fee, advise employees about how much to contribute and how to invest their contributions.
Some employers will be reluctant to move in this direction because they will appear to be paternalistic. But a significant and inescapable implication of this line of research is that employers cannot avoid responsibility in this area because there is no "neutral" plan design. Whatever design they choose will affect the retirement wealth of employees.
To wrap up, employers play an important role in their workers' financial security in later life and will continue to do so even with the shift away from defined benefit to defined contribution plans. An increasing amount of evidence indicates that how firms set up these plans will affect worker participation and contribution rates and what types of assets they will buy. As research continues to further our understanding of the connection between plan design and worker choices, changes can be instituted to promote financial security. For their part, workers have a responsibility to improve their financial literacy and develop the skills and confidence to practice strategies for effective financial management. This improvement of financial literacy will become even more crucial if individual accounts are introduced as a modification to the current Social Security system, as proposed by the Administration.
2. Deloitte and Touche, 2003 401(k) Annual Benchmarking Survey. (407 KB PDF) Return to text
3. Deloitte and Touche, 2003 401(k) Annual Benchmarking Survey. (407 KB PDF) Return to text
4. Nellie Liang and Scott Weisbenner (2001), "Investor Behavior and Purchase of Company Stock in 401(k) Plans--The Importance of Plan Design," FEDS Working Paper 2002-36 and NBER Working Paper W9131; James Choi, David Laibson, Brigitte Madrian, and Andrew Metrick (2001), "For Better or for Worse: Default Effects and 401(k) Savings Behavior," NBER Working Paper W8651. Return to text
6. Brigitte Madrian and Dennis Shea (2001), "The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior," Quarterly Journal of Economics. Return to text