Volatile States: Institutions, Policy, and the Performance of American State Economies
184
Volatile States: Institutions, Policy, and the Performance of American State Economies
184eBook
Available on Compatible NOOK devices, the free NOOK App and in My Digital Library.
Related collections and offers
Overview
Volatile States identifies institutions and policies that are key determinants of economic success and illustrates the considerable promise of a mean-variance criterion for assessing state economic performance. The mean-variance perspective amends applications of growth models that rely on the mobility of productive factors keyed to income levels alone. Simply measuring the level of growth in state economies reveals an incomplete and perhaps distorted picture of performance. Taking the volatility of state economies explicitly into account refines the whole notion of "economic success."
This book is essential reading for economists, political scientists, and policy-makers who routinely confront questions about the consequences of alternative institutional arrangements and economic policy choices.
W. Mark Crain is Professor of Economics and Research Associate, James M. Buchanan Center for Political Economy, George Mason University.
Product Details
| ISBN-13: | 9780472024070 |
|---|---|
| Publisher: | University of Michigan Press |
| Publication date: | 12/10/2009 |
| Sold by: | Barnes & Noble |
| Format: | eBook |
| Pages: | 184 |
| File size: | 6 MB |
About the Author
W. Mark Crain is Professor of Economics and Director of the Center for Study of Public Choice at George Mason University. Dr. Crain has authored more than a hundred articles and books in the fields of economics and politics. Most of these works focus on fiscal policy, government regulation, and the organization of legislatures. In addition to his academics pursuits, Professor Crain has served on eight state and local government boards and commissions during the administrations of four Virginia governors, including the state's Advisory Board of Economists. Dr. Crain has previously been on the faculties at UCLA and Virginia Tech, and he was Assistant to the Director of the U.S. Office of Management and Budget. Dr. Crain holds a Bachelor of Science degree in Economics from the University of Houston, and a Ph.D. in Economics from Texas A&M University.
Read an Excerpt
VOLATILE STATES
Institutions, Policy, and the Performance of American State Economies
By W. Mark Crain
THE UNIVERSITY OF MICHIGAN PRESS
Copyright © 2003
University of Michigan
All right reserved.
ISBN: 978-0-472-11303-3
Chapter One Champions of the State Growth League
The economic history of the United States in the twentieth century reflects enormous progress. Living standards more than quadrupled between 1929, the year of the infamous stock market crash and the beginning of the Great Depression, and 1999. Income per capita rose to just above $29,500 in 1999 from about $7,000 in 1929 (both denominated in 2000 dollars). Figure 1.1 shows the powerful upward trajectory in U.S. living standards over the course of these seven decades.
What dampens the optimism about future economic progress is that the U.S. economy grew at an increasingly slower pace in the closing decades of the twentieth century in comparison to earlier decades. Figure 1.2 illustrates this diminishing performance of the U.S. economy, breaking down the average annual growth rate by decade. In the depression-wracked 1930s real per capita incomes declined slightly, followed by the roaring 1940s, when real per capita income grew at an average annual rate of 3.8 percent. World War II clearly pulled the U.S. economy out of its Great Depression and fueled the best decade of growth in the twentieth century. In the postwar 1950s, growth moderated but remained at a healthy 2.6 percent annual rate before accelerating once again in the 1960s to 3.3 percent. Then came the slowdown: growth rates fell to 1.9 percent in the 1970s, to 1.6 percent in the 1980s, and even lower to 1.3 percent in the 1990s.
A rigorous analysis of the U.S. income data during these seven decades confirms this thesis of a national slowdown. Estimating the regression model specified in equation (1.1) over various subperiods provides a basis for identifying significant differences, both in statistics and in magnitude, in the national income growth rates.
ln (Real Income per [Capita.sub.t]) = constant + s ([Trend.sub.t]) [[epsilon].sub.t], (1.1)
where ln stands for the natural logarithm, [Trend.sub.t] stands for a linear time trend, and [[epsilon].sub.t] represents the regression error term. Using this regression model, the estimated coefficient for s reflects the annual growth rate in real per capita income in a given time period. The results of estimating equation (1.1) for the 1929 to 1999 time period and two subperiods, 1950 to 1999 and 1970 to 1999, are shown in table 1.1. The relevant results from the regression analysis in table 1.1 that test for growth rate declines are summarized in table 1.2. As shown in the first row, real income per capita grew annually by 3.0 percent between 1929 and 1969. This growth rate equals exactly twice the 1.5 percent annual growth rate for the 1970 to 1999 period (the difference is significant at the 1 percent confidence level). The estimated growth rate for the period 1950 to 1969 is 2.6 percent, which is also greater than the 1.5 percent growth rate in the 1970 to 1999 period (this difference is again significant at the 1 percent confidence level). The second row of table 1.2 computes the relative differences in economic growth during these three epochs. The annual growth rate was 100 percent higher in the 1929 to 1969 period compared to the 1970 to 1999 period and was 73 percent higher in the 1950 to 1969 period compared to the 1970 to 1999 period. In short, while the U.S. living standards continued to grow in the last three decades of the twentieth century, this growth slowed significantly compared to the preceding decades.
As a final gauge of the U.S. economic slowdown, table 1.3 presents a simple regression that estimates the growth rate in the last half of the twentieth century. Here the dependent variable is the annual growth rate (continuously compounded), and the dummy variable breaks the sample at its midpoint, equal to zero for the years 1950 through 1974 and equal to one for the years 1975 through 1999. The constant term, 0.027, indicates a growth rate of 2.7 percent in the 1950-74 period, and the dummy variable indicates that the growth rate dropped by 1.2 percentage points in the 1975-99 period. This estimated coefficient on the dummy variable is significant at the 5 percent level, rejecting the thesis that economic growth in the United States remained unchanged over the 50 years.
The Deceptive Big Picture
These various metrics confirm the conventional wisdom: the aggregate growth rate for the United States began a steady decline in the 1970s that shows no sign of abating at the turn of the twenty-first century. What is less well known is how accurately this aggregate slowdown in U.S. growth characterizes the economic experience of the individual American states. When the model used for table 1.3 is estimated for each of the 50 state economies for the 1950 to 1999 period, an interesting pattern emerges. In only 13 of the states (roughly 1 in 4) do we And a statistically significant decline in the income growth rate using the 5 percent confidence level as the cutoff criterion. If we use a less stringent 10 percent confidence level as the cutoff criterion, we still And a statistically significant drop in economic growth in only 22 states (less than half the nation).4 Growth rates showed no statistically significant decline in more than half of the states. To put these results another way, the economies of 28 American states departed from the "national" pattern and showed no significant slowdown in the last half of the twentieth century.
With this initial piece of empirical analysis we begin to dispel the notion of the "U.S. economy." The big picture provided by the aggregate data fails to represent adequately the experiences of individual states, and in some cases, as we shall see, the aggregate U.S. data grossly misrepresent state-level circumstances. That major subnational pockets depart from the presumed national picture suggests the importance of state-specific determinants of economic performance.
Traditional Measures of State Economic Performance
To pursue this point further, consider the dissimilarity in economic performance among the American states in the last three decades of the twentieth century. Figure 1.3 ranks the 50 states based on growth rates for the 1969-99 period. The values reflect the average annual growth in real income per capita (using the continuously compounded method described in note 1).
As a benchmark, in the 1970s through the 1990s income per capita for the total United States grew at a 1.63 percent rate using this measurement method, which is close to the growth rate in the median state of 1.69 percent. (Given an even number of American states, Nebraska and Kansas both claim the median position and rank as twenty-fifth and twenty-sixth in Ag. 1.3.) North Carolina sits on top of the distribution with an annual growth rate of 2.15 percent. Alaska comes in dead last with a 0.94 percent rate, less than half of the growth rate in North Carolina. Figure 1.3 indicates that many of the fastest growing states tended to be in the South; six of the ten fastest growing states were southern. The slowest growing states tended to be in the West and Midwest.
A casual glance at figure 1.3 shows that state growth rates for the most part ranged between 1.3 percent and 2 percent. This visual inspection comes quite close to a statistical characterization of the data. The range in growth rates represented by the mean value plus and minus one standard deviation is 1.4 percent to 2 percent; two-thirds of the states lie within this range of growth rates. Growth rate differences of less than 1 percentage point might appear trivial, hardly worth notice. Yet, seemingly small growth rate differences yield major economic consequences over time. As a demonstration, compare the historical path of living standards in Missouri and Virginia over these three decades. In 1969, Missouri and Virginia had virtually identical living standards, and both states nearly matched the median income across states in that year ($16,800 in 2000 dollars). Subsequently, Virginia grew at a 2.04 percent annual rate and Missouri grew at a 1.64 annual rate, a difference of 0.4 percentage points. How did their living standards compare 30 years later? In 1999, per capita income rose to $30,809 in Virginia and to $27,196 in Missouri (again in 2000 dollars). Thus, starting with income parity in the late 1960s, Virginia income exceeded Missouri income by $3,500, or 13 percent, in 1999. The lesson, of course, is that seemingly small differences in growth rates pile up and over the years result in meaningful differences in the well-being of the typical family.
Table 1.4 presents state growth estimates using two alternative measurement techniques, the least squares method and the continuously compounded method. The table orders the states from fastest growth (equal to 1) to slowest growth (equal to 50). For comparison, table 1.4 also lists the continuously compounded growth rates (the same values plotted in Ag. 1.3) and the state rankings based on that value.
Using the least squares method New Hampshire has the highest growth rate (2.14 percent annually) followed by Connecticut and Massachusetts (2.11 percent). Note that the frequent assertion that the economies of the southern states lead the nation depends upon which growth measure is used. New England states claim the top three spots using the least squares method. In addition, the economic performance of other New England and Mid-Atlantic states ranks considerably higher using the least squares method compared to the continuously compounded method. New Jersey moves into the number 10 spot, and Rhode Island moves into the number 11 spot, whereas these two states appeared near the middle of the pack based on the compound growth metric. New York and Delaware also improve their relative performance rankings substantially. For the United States as a whole income per capita grew at a 1.52 percent rate using the least squares method, right at the growth rate in the median states of 1.51 percent (South Dakota and Missouri). Alaska again finishes last with a 0.33 percent growth rate, which is 85 percent less than New Hampshire's growth rate. Overall, the simple correlation coefficient between the continuously compounded growth rate and the least squares growth rate is 0.86; in other words the two measures are closely, but not perfectly, related.
Figure 1.4 depicts the extensive variation among state economies during the 1969-99 period. This figure provides a frequency distribution, which divides income per capita growth rates into bands of 25 basis points. For example, seven states grew at an annual rate between 1 percent and 1.24 percent, and six states grew at an annual rate between 2 percent and 2.24 percent.
A slightly different way to assess economic performance is to gauge the change in income relative to the change in the state's work force. Figure 1.5 ranks the 50 states based on this indicator. There the values reflect the average annual growth in real income per worker growth rates for the 1969-99 period (using the continuously compounded method described in note 1). As a benchmark, in the 1970s through the 1990s income per worker for the entire United States grew at a 0.67 percent rate using this measurement method, which exactly equals the median growth rate per worker among states. (Missouri and Illinois occupy the median position in Ag. 1.5.) North Carolina returns as the growth champion, with an annual growth rate of 1.28 percent. Alaska Anally moves out of the cellar, replaced by Montana, which experienced an annual growth rate in income per worker of 0.04 percent. Figure 1.5 places eight southern states among the top ten in terms of income per worker growth; all of the bottom ten states are from the Midwest and West.
Table 1.5 ranks the states based on growth in real income per worker. Based on the least squares method, North Carolina, the leader, has a growth rate of 1.12 percent, followed by South Carolina (1.09 percent) and Georgia (1.03 percent). Five states experienced negative growth in income per worker: North Dakota, Alaska, Montana, Iowa, and Utah. Income per worker remained unchanged in Wyoming over the three decades.
It is important to note that the growth rates in income per worker run much lower than the growth rates in income per capita. The median growth rate in income per worker is 0.43 percent, compared to the 1.51 percent median growth rate in income per capita. This difference reflects the fact that labor force growth outstripped population growth by a wide margin. The rise of the proportion of women who worked outside the home during these three decades accounts for most of this demographic shift. Also noteworthy are some significant changes in how particular states' rankings differ under the alternative growth metrics. Mississippi provides a telling example. Gauging performance in terms of income per worker growth, Mississippi ranks fifth (using the least squares method in table 1.5), whereas it ranks twelfth in terms of income per capita growth (table 1.4). Florida ranks a respectable fourteenth in terms of per worker growth, whereas it ranks a mediocre twentieth in terms of per capita growth. California ranks twenty-ninth in per worker growth, fairly near the median value, whereas it ranks forty-fifth in terms of per capita growth. Finally, with regard to the income per worker measures, the simple correlation between the continuously compounded growth rate and the least squares growth rate is 0.95, indicating that these two measures yield quite consistent performance rankings.
The Dispersion of State Living Standards
The analysis now turns to a second performance measure, the level of real income, which is the traditional indicator used to assess living standards. First, table 1.6 ranks the states based on the level of real income per capita. Per capita incomes and state rankings are shown for two four-year intervals, 1996-99 and 1969-72. Here the analysis uses the four-year average for each state at the beginning and the end of the period to dampen the importance of a random downturn or upturn that may have occurred in a single year. However, comparing the state income rankings in 1999 to 1969 yields basically similar results.
At the close of the twentieth century (the 1996-99 period), real per capita income in the median state (Oregon and Kansas) equaled $26,773 and ranged from $38,964 in Connecticut to $20,647 in Mississippi, a spread of almost two to one. Alaska topped the rankings and Mississippi bottomed the rankings in the 1969-72 period, and there again the income spread was about two to one. While the proportionate distance between the richest and poorest states remained about the same, the pecking order in living standards was substantially reshuffled. Starting at the extremes, among the ten states at the bottom of the income ladder in the 1969-72 period, four had climbed out by the 1996-99 period (South Carolina, Tennessee, North Carolina, and South Dakota). Among the ten richest states in the 1969-72 period, four had fallen out of this coveted category by the end of the twentieth century (Alaska, Hawaii, California, and Delaware).
Table 1.7 further displays the considerable dispersion in the state incomes, assessed this time in terms of the level of real income per worker. These results follow the format used in table 1.6, reporting the mean per worker income levels and state rankings at the beginning and end of the three decades. Based on this indicator of living standards, New Jersey and Connecticut topped the rankings at the beginning and at the end of these three decades. North Dakota and Montana ranked lowest in the 1996-99 period, replacing South Carolina and Mississippi, which had the lowest incomes per worker in the 1969-72 period. Among the poorest ten states by this ranking, seven changed between 1969-72 and 1996-99. Among the ten richest states, three fell from grace during these three decades.
Figure 1.6 provides a visual look at these relative state income data using income per capita as the performance indicator. The layout in figure 1.6 indicates the extent to which the relative well-being among the states remained constant over the 1969-99 period. The states' rankings in terms of income per capita in the late 1960s to early 1970s period are shown on the vertical axis, and the states' rankings in the late 1990s are shown on the horizontal axis. The richest state ranks as 1, and the poorest state ranks as 50. This setup easily identifies those states whose relative economic conditions remained unchanged over these three decades; such states will fall along the 45 degree line in figure 1.6. For example, Mississippi ranks as the poorest state throughout the three decades. Only three other states maintained the same relative income per capita ranking: Rhode Island, Pennsylvania, and Florida.
(Continues...)
Excerpted from VOLATILE STATES by W. Mark Crain
Copyright © 2003 by University of Michigan . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.