Issue 5, September/October 1999
Federal Reserve Bank of Dallas

The Economics of Prosperity: A Texas Tale

Like much of the rest of the nation, Texas is enjoying a low-inflation, low-unemployment economy. The Texas misery index—the sum of the inflation and unemployment rates—is bouncing near its 30-year low (Chart 1), and both its components are lower than they were during the early-1980s boom. Unemployment is below 4 percent in more than half the state and, for the first time in recent memory, in single digits along much of the border.

In a low-unemployment environment, labor force growth limits employment growth, and barring a major change in the percentage of the population seeking work, population growth limits labor force growth. Therefore, the patterns of unemployment, labor force participation and population growth will heavily influence the economic future of Texas. This article explores these patterns and discusses their implications.

Unemployment
Unemployment rates in Texas vary widely (Chart 2). For example, the unemployment rate is more than seven times as high in McAllen as it is in Bryan/College Station. Unemployment tends to be higher along the coast and on the border with Mexico and lower in North and Central Texas. Interestingly, three of the four cities with the lowest rates—Austin, Bryan/College Station and Lubbock—are also home to major state universities.

While levels vary substantially, there is definitely a common trend in Texas unemployment rates. As Chart 2 shows, unemployment has been falling throughout the state. Over the past year, rates have fallen everywhere except oil-sensitive cities like Houston and Midland/Odessa (and Bryan/College Station, where there was essentially no room for further declines). The decreases have been particularly sharp in Brownsville, McAllen and Texarkana, where unemployment fell more than 2.5 percentage points between July 1998 and July 1999. Unemployment rates in Dallas, Fort Worth, Killeen, San Antonio, Sherman and Waco are now less than half what they were when rates began falling seven years ago. Unemployment in Dallas and Fort Worth hasn't been lower in 20 years.

Labor Force Participation
Labor force participation also varies dramatically across Texas. The civilian participation rate is the share of the working-age population (that is, everyone over 16) that is working or actively seeking work. It excludes people who are in the military, retired, attending school full time, keeping house or staying at home with the kids. Labor force participation rates tend to be highest in communities with relatively few people of retirement age or children in need of parental supervision and lowest in areas with low real wages and high unemployment.

Chart 3 illustrates deviations from the national average participation rate of 67 percent. As the chart shows, rates are already quite high in much of the state, particularly in areas with low unemployment. (The major exception is Killeen, where Fort Hood skews the data.) The participation rates for Austin and Dallas are more than 10 percentage points above the national average. Among major U.S. cities, only Minneapolis/St. Paul has a higher rate than Dallas/Fort Worth.

Overall, Texas participation rates have been drifting upward in metropolitan areas with low unemployment and drifting downward in areas with high unemployment; beyond that, there has been little meaningful change among the metros. The Texas areas with the highest rates in 1998—Austin, Dallas, Fort Worth and Houston—also had the highest when unemployment rates began falling in 1992. Among major Texas metropolitan areas, only Austin has seen a substantial increase in its participation rate over this period of tightening labor markets. Austin's rate rose from 75 percent in 1992 to 81 percent in 1998.

Population Growth
The Texas population tends to grow at twice the national rate. Two important factors explain this pace—a faster rate of natural increase (meaning that the young Texas population produces substantially more births than deaths each year) and strong net domestic migration (meaning that more people from elsewhere in the country move in than Texans move out). However, as Chart 4 shows, there is at least as much variation in Texas' population growth rates as there is in its unemployment and labor force participation rates.

The working-age population is growing most rapidly in Laredo, McAllen, Austin and Dallas. Laredo and McAllen benefit from especially strong rates of natural increase and international migration; on net, domestic migration has a negligible effect on these cities. In contrast, Dallas and Austin grow more rapidly than much of the rest of the state because net domestic migration is so strong. In 1998, 51,000 people moved into these two metro areas from elsewhere in Texas and the United States.

At the other end of the spectrum, domestic migration was the primary source of drag on the weakest Texas metros. The areas shown in brown in Chart 4 lost population to other parts of the state and the nation in 1998. Interestingly, no Texas metro area lost population to international migration in 1998.

Implications
All tight labor markets experience the same economic forces, albeit to varying degrees. Therefore, focusing on one or two can illustrate the broader economic implications for the state as a whole.

Austin and Dallas have by far the tightest labor markets in Texas. Unemployment rates are low, and labor force participation is unusually high. As a consequence, there are nearly nine jobs for every 10 residents between the ages of 16 and 65 in Austin and Dallas. Meanwhile, population growth has not kept up with recent job growth (nonfarm employment in both areas has increased by at least 4 percent a year for the past three years). Something's got to give. Because there will always be some "frictional unemployment," as workers search between jobs or gather information upon entering the labor force, there is little room for unemployment rates to fall further. Therefore, the current rate of job growth in Austin and Dallas is unsustainable without a significant increase in either labor force participation or net migration.

The market forces needed to lure workers into the Austin and Dallas labor forces will induce a number of changes. First, there will be significant upward pressure on labor compensation. As many employers find themselves chasing the same set of workers, bidding wars will erupt for workers with specific skills. A recent Manpower survey indicated that one-fifth of Dallas employers were planning to hire in the fourth quarter. Some of them had best prepare for sticker shock. It's becoming a seller's market for labor in Austin and Dallas.

Increasing labor compensation may not take the form of rising wages, however. Industry contacts suggest that working environment, fringe benefits and stock options are becoming an increasingly important part of the total compensation package.

Higher compensation should increase labor force participation, but the near-term effect is likely to be modest. Participation rates tend to change at a glacial pace, Austin's recent experience notwithstanding. For example, the Texas rate has changed less than 1 percentage point over the course of the decade. Simple diminishing returns will keep Austin from continuing to increase its participation rate at the pace of the past eight years.

Higher compensation is more likely to attract economic migrants than to draw existing residents out of the woodwork. Therefore, tightening labor markets in Austin and Dallas could increase the rate of net domestic migration into the two areas. Such a change would only reinforce an existing trend; as Chart 5 shows, the Texas metros with the tightest labor markets experienced the greatest net domestic migration in 1998.

On the other hand, as economic conditions have improved nationwide, the factors that were pushing workers out of other states have dissipated, and any influx of workers would bid up housing costs and push up the cost of living in Austin and Dallas. These factors could counterbalance the attraction of wage increases. So unless labor compensation rises dramatically, net migration into the two areas is unlikely to accelerate markedly.

It is more likely that tight labor markets in the two areas will attract commuters from the surrounding counties. Such a pattern is particularly likely in Dallas. There are two yardsticks by which metro area employment is measured: by the location of the worker and by the location of the firm that employs the workers. Usually, the worker-based measure of household employment produces a higher job count because it includes self-employed and agricultural workers who are not captured by the establishment survey. Since 1997, however, the Dallas establishment survey has reported more jobs than the household survey. This shift could arise from a number of factors, but it most likely reflects Dallas firms' hiring of an increasing number of non-Dallas residents (who are not included in the household survey estimates for the area). If the commuting becomes common enough, the boundaries of the metropolitan areas will be expanded after the 2000 census to sweep up the outlying counties and reflect the new economic reality.[1]

While rising wages will pull some people out of school or retirement and others out of an adjacent county, the supply side is only part of the market response to tight labor markets. Firms are the other side of the equation, and they are as likely to move as workers. Firms often cite the availability of workers with the appropriate skills as a major factor in their location decisions. If firms cannot expand easily or must pay a wage premium to expand in Austin or Dallas, they will expand elsewhere instead. Some of those alternative locations will be in Texas, but not all. For example, tight labor markets were cited as one of the important factors behind Dell Computer Corp.'s recent decision to build its first major non-Austin facility—in Tennessee. Thus, even as good economic times continue, job growth is likely to slow significantly in Austin and Dallas.

— Lori L. Taylor

About the Author Taylor is a senior economist and policy advisor in the Research Department at the Federal Reserve Bank of Dallas. Note An outlying county is included in a metropolitan statistical area (MSA) on the basis of commuting patterns and the urbanicity and population density of the outlying county. Generally, counties are not added to MSAs between censuses unless the central city expands into the county (through annexation, for example).

Where Have All the Savings Gone?

Ever since Ben Franklin wrote "A penny saved is a penny earned," Americans have been taught that saving is a virtue.[1] Having accepted this principle, many economic observers are concerned about the recent sharp decline in America's personal saving rate. Many economists are also concerned because they believe personal saving is a requisite for economic growth and progress. Such progress requires a steady stream of investment expenditures for the development of new technologies and for the purchase of new plant and equipment. To generate this investment stream, society must forgo current consumption so resources can be diverted from the production of consumer goods to the production of capital, or investment, goods. Saving, then, is the means by which resources are diverted from current consumption to future growth.

As can be seen in Chart 1, the personal saving rate has moved irregularly downward since 1980 and by 1998 was close to zero. The Bureau of Economic Analysis (BEA) rate actually dropped below zero in 1998 and has remained negative in 1999.[2]

The near-zero and negative monthly personal saving rates for 1998 and 1999 represent a dramatic break with the past. Monthly saving rates in the late 1970s and early 1980s generally oscillated between 6 percent and 10 percent, with a spike up to 13.6 percent in 1980 (Federal Reserve series). Since the early 1980s, however, the rate of personal saving has shown a marked decline, interrupted only by a modest recovery between 1989 and 1992. The average monthly saving rate for 1988-91 (5.5 percent) was one-fourth lower than that for 1975-81 (7.2 percent). More recently, the 1995-98 rate (2 percent) was only about one-fourth that of 1975–81.

The persistent decline in the personal saving rate seems paradoxical, as American living standards have been steadily improving and the nation's stock indexes rising.[3] Commentators have sought to explain this phenomenon by pointing to policy decisions or the economic trends of the past two decades. Tax rate increases adopted in 1990 and 1993 and the rising trade deficit have been popular targets. Some economists speak of a change in the very nature of Americans—from Ben Franklin-like good citizens who see saving as a virtue to profligate consumers who see conspicuous consumption and even excess debt as privileges of an advanced economy infected with "luxury fever."[4] Both the current administration and Congress have proposed legislation to address America's alleged inadequate saving rate. It is now a virtual media pastime to bemoan the nation's profligacy and the problems our current "consumption-binge" mentality is bound to create for future generations.

Should we worry about the saving rate trend? If today's saving behavior is a rational, healthy response to economic conditions, we can ignore the rhetoric about approaching disaster. When one looks at the entire economic picture and employs better indicators of the consumption/saving trade-off than the simple personal saving rate, the often-invoked "savings crisis" disappears. This is important because it means we can stop fretting over whether economic growth will suffer and whether Americans will have sufficient resources for their futures.

Why Saving Is Higher Than It Appears
To save is to postpone consumption. A nation saves when a portion of current output is not consumed today but set aside for the future as either finished goods or capital investment. Actually, America's personal saving might be higher than it appears in Chart 1 because the chart does not include all forms of saving (nonconsumption). The personal saving rate is derived by dividing personal savings of all Americans by their aggregate personal disposable income. But these terms do not mean what most Americans might think because personal saving is not calculated by adding up the various saving instruments of the population. On the contrary, the personal saving rate is an accounting construct calculated by subtracting personal consumption expenditures from personal disposable income (the latter being personal income less taxes), then dividing the result by personal disposable income. Derived in this manner, the personal saving rate does not include corporate saving, the accumulation of consumer durables or human capital expenditures.

Chart 2 illustrates the effects of including these related economic magnitudes in private sector saving. The chart adds to personal saving the net accumulation of consumer durables, undistributed corporate profits—which the BEA includes in private saving but not in personal saving—and human capital investment as measured by personal education expenditures.[5] Not surprisingly, this chart gives a brighter picture of what Americans are doing with their incomes. As Chart 3 shows, they are currently saving at an annual rate of about 10.25 percent of their personal income.[6]

People do not save for the sake of saving. They save to spread consumption over their lives. It is interesting to note, then, that when they purchase durable goods or education, the official saving rate falls. In fact, Americans' spending on durables and education is rising faster than income. Certainly, some of these expenditures may not prove effective in providing for future consumption, and our savings definition is open to criticism on those grounds. Nevertheless, these additions need to be carefully considered before drawing the conclusion that the savings sky is falling.

Net Worth: The Missing Variable?
Perhaps personal saving isn't even the right statistic to analyze when seeking to understand America's consumption/investment trade-offs. Americans save by accumulating a portfolio of assets, some financial and some nonfinancial (durables and education expenditures, as previously noted). If the value of Americans' total portfolio rises, their net worth rises and less immediate saving is required. In fact, we ought to see an inverse relationship between what the Commerce Department calls personal saving and overall net worth, and we do. Chart 4 shows real net worth rising at a record rate since the mid-1980s.

The value of stock portfolios rose from $7.2 trillion in 1996 to $10.8 trillion in 1998, a staggering 50 percent increase in just two years. And the equities market has continued to climb to new records in 1999. The present net worth of all U.S. households is $36.8 trillion, almost double the 1996 combined GDPs of the world's five largest economies—the United States, Germany, France, Great Britain and Japan. At the same time, according to the Federal Reserve's funds flow report, consumer debt has grown more slowly than asset appreciation.

Americans are taking on more debt because they can afford to. Chart 5 shows that households hold more than six times their current incomes as net assets. Not surprisingly, as Chart 6 clearly shows, they have increased their consumption, and their ability to spend comfortably, as their net worth has risen. As opportunity, stability, low unemployment and economic growth have become the new American economic norm, the simpler "saving or consumption" world has become obsolete. For this reason, we should not expect participants in an evolving, national market economy to save, year after year, some predictable, constant percentage of their income.

As the nation's wealth, demographic makeup and economic opportunities change, so might the personal saving rate. What we have shown thus far is that when a definition of asset accumulation more comprehensive than "personal saving" is used, the so-called savings crisis largely disappears. Americans are spending today as if they believe that not only is there a tomorrow, but it's going to be a very good one.

Some Policy Considerations
No economist or government agency knows the economically optimal allocation between current and future consumption. Only individuals can make such choices, and they do so based on their goals, means, expectations and incentives. Even though U.S. private saving has declined less than critics claim—and asset accumulation not at all—it may still be desirable for Americans to save more to stimulate private investment and capital formation. Americans now face a number of disincentives to save. Several current government policies discourage saving. Some possible changes that would increase saving are as follows:

• Tax consumption, not income. Taxing income only when spent—not when saved—would encourage private saving and asset accumulation. Under certain assumptions, equivalent results could be achieved by eliminating the tax on capital income, such as dividends, interest and capital gains. Either of these reforms would eliminate the double tax currently imposed on savers.
   
• Reduce or eliminate the corporate income tax. Short of eliminating tax on all capital income, repeal of the corporate income tax would reduce the overly burdensome tax on saving and investment in U.S. business. Investors in U.S. corporations currently pay three taxes—one when the money is earned, one when the business earns a profit (the corporate profits tax) and one when the dividends are paid out to shareholders. Saving and investment thus suffer.
   
• Reduce or eliminate the "death" tax. The estate and gifts tax has become increasingly onerous in recent years as markets have lifted Americans' wealth above the untaxed household ceiling (currently $650,000 and rising to $1 million in 2006). Eliminating this tax would encourage private saving, especially lifetime wealth accumulated in family-owned businesses and farms, which under current law often must be sold to pay the tax.
   
• Simplify and stabilize the tax code. A small, simple and predictable tax is best for stimulating economic activity, including saving. When the tax code is difficult to understand and interpret, or subject to frequent and extensive revision, private saving suffers.
   
• Reform the federal bankruptcy code. Generous federal bankruptcy laws encourage citizens to spend and borrow without consequence. Tightening the laws would encourage Americans to accumulate wealth, not debt.

Conclusion
The general query "Is America saving enough?" is probably not answerable. For years, many policy commentators have warned that frugal Japan would someday overtake America as the world's premier economic power. That was before the Japanese economy sank, many of its larger banks encountered financial difficulties, and its stock and real estate markets collapsed. Japan's high national saving rate did not prevent economic turmoil, nor is it helping Japan overcome it. What policy advice has Japan received from the same commentators who decry America's profligate ways? Consume more and save less!

It has probably always been the case that some people save too much and others save too little, at least from the perspective of third-party observers. But since individuals differ in their goals, it is problematic to evaluate the saving of an entire nation. In view of the arguments presented here, though, it is clear that pessimism regarding Americans' saving is largely unfounded.

We should remember that our national income accounting definitions were created in another era—one dominated by physically countable manufactured and agricultural output. Today, information and services are the twin pillars on which the growth and prosperity of our economy rest. It does us little good to continue attempting to navigate tricky public policy shoals with antiquated national income and product accounts gauges. As our economy and economic theories change, so must our methods of measurement. Only then can we hope to accurately judge whether Americans are saving too little. . .or too much.

— Robert L. Formaini and Richard B. McKenzie

About the Authors Formaini is a senior economist in the Research Department of the Federal Reserve Bank of Dallas. McKenzie is a professor in the Graduate School of Management at the University of California, Irvine. Notes The authors thank Mike Cox, Jason Saving and Alan Viard for their valuable input, and Justin Marion and Kathryn Cook for research support. Old Ben understated his case. A 22-year-old who saves a penny and receives the average rate of return of the S&P 500 across the intervening years will have 32 pennies when he retires at age 67.   On September 8 of this year, the Commerce Department's Bureau of Economic Analysis announced it has decided to revise the calculation, retroactively to 1929, of several macroeconomic variables, including the personal saving rate. Government workers' pension contributions will now be counted as personal, rather than government, saving. While this does not change GDP, it does increase the personal saving rate by an estimated 1.5 percent to 2 percent, or about $100 billion in the 1990s alone.   W. Michael Cox and Richard Alm, Myths of Rich and Poor: Why We're Better Off Than We Think (New York: Basic Books, 1999).   Robert H. Frank, Luxury Fever: Why Money Fails to Satisfy in an Era of Excess (New York: Free Press, 1999). The New York Times agrees: Stephen Roach, "Spending Ourselves into Oblivion," December 11, 1998, p. 35.   The net accumulation of consumer durables taken from BEA data represents purchases less depreciation. For human capital expenditures, no official data series exists to use as a basis on which we could reliably measure and subtract depreciation. Also, we have revised only the private side of saving, ignoring the upward trend in government saving. Federal, state and local government surpluses make up part of national saving and must be considered before making judgments about a "savings crisis." Just prior to publication, we became aware of similar work by William Gale and John Sabelhaus ("Perspectives on the Household Saving Rate," Brookings Papers on Economic Activity, no. 1, 1999, pp. 181–224), who reach similar conclusions, although we were working independently. Although their revised savings definition is not the same as ours, they estimate about a 2 percent decline in saving during 1975–98, consistent with what we found.   The ratio we use in Chart 3, personal savings and related items/personal disposable income and undistributed corporate profits, has been relatively stable since 1970, peaking at 17 percent in 1973 and moving slightly downward during the following decade but never varying during that decade by more than 2 percent. To avoid artificially increasing the ratio, we add undistributed corporate profits to the denominator as well as to the numerator.

Beyond the Border
The World's Newest Currency

January 1 marked the formal launch of Economic and Monetary Union (EMU) in Europe as 11 nations of the European Union (EU) merged their currencies into a new single currency, the euro, and ceded sovereignty over monetary policy to a new supranational institution, the European Central Bank (ECB). The 11-nation entity that shares this new currency is similar in size (economically) to the United States, automatically making the euro the world's second most important currency after the dollar.

The launch of the euro and the creation of the ECB to manage it are part of the longer-term process of European integration that began shortly after World War II. The euro's creation eliminates all exchange rate risk between the participating nations and will further deepen the single market in goods and services that has existed since 1992. By ceding sovereignty over monetary policy to the ECB, individual countries can no longer tailor their monetary policies to domestic economic conditions, but must instead accept the policy set by the ECB on the basis of economic conditions in the euro area as a whole.

It is too early to say for sure whether EMU will succeed and whether the euro will be a strong currency. Many economists have raised serious doubts about the wisdom of EMU and have argued that the project is doomed to fail sooner or later. Periods of boom in some countries that are accompanied by periods of recession in others, in conjunction with limited labor mobility, will create strains that will test the new institutions. The skeptics point to historical evidence that monetary unions that are not accompanied by political unions invariably fail. The optimists point to the important role shared currencies play in building a common political identity and note that EMU differs in many important respects from earlier attempts at monetary union.

The fact that EMU membership goes hand in hand with access to the single market makes the costs of seceding from the monetary union quite high. Also, the creation of a single institution (the ECB) to manage the new currency, and the accountability of the ECB to the European Parliament, should enhance the legitimacy and durability of EMU.

It is also too soon to tell whether the euro will prove to be a strong currency and a worthy successor to the perennially strong Deutsche mark. Insofar as it is possible to design a currency to be strong, the euro has a lot going for it. Current conventional wisdom holds that central bank independence is a key prerequisite of a strong currency. The ECB is probably the most independent central bank in the world, and it has an unambiguous mandate for the pursuit of price stability. The ECB is only obliged to support the other policies of the EU to the extent that this support does not compromise its primary objective of price stability. But the ECB will be entering uncharted territory when it comes to building a constituency for its policies across national borders. The ability of Deutsche Bundesbank to pursue an independent monetary policy was due in no small part to the popular support it enjoyed among the German electorate.

Much was made earlier this year of the euro's steady fall against the dollar on foreign exchange markets (Chart 1). To some this was a sign of inherent weakness in the new currency. However, a number of factors should be considered before reading too much into short-term movements in the dollar–euro exchange rate. First, at least part of the decline was simply an unwinding of the appreciation of the legacy currencies against the dollar in the latter half of 1998. As relative growth prospects in the United States and the EU shifted, so too did the exchange rate. Second, uncertainty about the direction of economic policy in Germany, the euro area's largest economy, and confusion about the ECB's attitude toward exchange rate developments detracted from market confidence in the first months of the new currency's existence.

The exchange rate has been more stable in the past couple of months, as the outlook for growth in Europe has improved. There is growing evidence that the slowdown in economic activity in Europe that began late last year has come to an end and that the continent is poised for faster growth in the coming year. Both Germany and Italy show signs of an upturn in economic activity, while the countries on the periphery (Finland, Ireland, Spain and Portugal) continue to grow rapidly. The ECB's first real test will occur when it comes time to raise interest rates to stave off incipient inflation.

— Mark A. Wynne

About the Author Wynne is a senior economist and research officer in the Research Department at the Federal Reserve Bank of Dallas.

Regional Update
Focus on the Energy Industry

Two years ago, Texas energy producers were going like gangbusters. Futures markets predicted the price of West Texas Intermediate crude would stay near $20 per barrel for the foreseeable future. Industry contacts reported shortages of rigs and personnel. There was a 12- to 18-month backlog for drill pipe.

Four months later, the price dropped below $19 per barrel and stayed there for nearly two years. In December 1998, as producers were setting their exploration budgets for 1999, the price of West Texas Intermediate hit a 12-year low of less than $11 per barrel.

Today, although the price of oil is again well above $20 per barrel, drilling activity has not recovered. Extraction employment is showing hints of renewal but remains 16,500 jobs below its year-ago level. The Texas rig count has changed direction recently but has only climbed back to its August 1998 level (see chart below). It will need to increase by 40 percent to reach the level it attained the last time prices were this good.

The cautious industry response probably reflects a belief that recent price increases are a temporary windfall. The futures market expects prices to fall to the $18–$19 range next year. However, even with the declines, oil prices are still expected to exceed the levels on which the drilling budgets were based and remain well above the December 1998 level.

— Lori L. Taylor

About Southwest Economy Southwest Economy is published six times annually by the Federal Reserve Bank of Dallas. The views expressed are those of the authors and should not be attributed to the Federal Reserve Bank of Dallas or the Federal Reserve System. Articles may be reprinted on the condition that the source is credited and a copy is provided to the Research Department of the Federal Reserve Bank of Dallas. Southwest Economy is available free of charge by writing the Public Affairs Department, Federal Reserve Bank of Dallas, P.O. Box 655906, Dallas, TX 75265-5906, or by telephoning (214) 922-5254.