By Pavlina R. Tcherneva

Associate Director for Economic Analysis

 

For most of the 1990s the unemployment rate in the state of Missouri (MO) has been noticeably lower than the national unemployment rate (Figure 1).  And although it exhibits significant variability, it has averaged well below the national trend. During the latest downturn, however, Missouri experienced a significant increase of unemployment much earlier than the nation as a whole, and it has yet to show signs of recovery.  In fact, in June 2002 the state registered the steepest annual drop in employment in the country.  Furthermore, since May 2004, MO’s unemployment has not only outpaced the national rate but has also continued to trend upward.  What this data suggests is that, although during the 1990s goldilocks economy Missouri workers fared better when compared to those in the U.S. as a whole, they have also suffered disproportionately more from the downturn.

 

 

 

 

 

 

 

 

 

 

           

 

What has this increase in unemployment meant for the state of Missouri and its residents and is it possible to quantify its economic impact?  This special report provides one answer to these questions by computing the cost of unemployment.

The Cost of Unemployment

Measures of the impact of unemployment are generally concerned with two types of costs – direct and indirect.  The direct costs of unemployment gauge the actuarial costs to the state of paying unemployment insurance (UI) and other welfare benefits to jobless recipients.[1]  Of course an increase in unemployment does not just mean an increase in state outlays in terms of UI and other benefits.  It also means real loss of income and output for the economy as a whole.  These are generally captured when calculating the indirect costs of unemployment.

Indirect costs are two types: economic and social costs.  Measures of the economic costs of unemployment aim to capture the difference between potential and actual output.  Had the jobless been gainfully employed, they would have contributed to the production of goods and services in the economy and, therefore, to growth.[2]  This loss in output is usually called the economic (or opportunity) cost of unemployment.  In addition to the foregone output, the economic indirect costs also include the costs of increased poverty and income inequality.

Loss of output, however, is too narrow a measure of the cost of unemployment, because the unemployed are not simply unutilized factors of production.  Unemployment brings numerous other non-pecuniary costs, which are generally called the indirect social costs of unemployment.[3]  Unemployment results not just in the loss of income, but also in the deterioration of human capital and employability.  It is also linked to an increase in certain crimes and health costs.  It fosters social antagonism, exclusion, alienation, and stigmatization (Forstater, 2003).

 The calculation of the cost of unemployment is a complicated matter for two main reasons.  Most importantly, it is difficult (if not impossible) to assign a monetary value to the indirect costs of unemployment (both economic and social).  Secondly, while the direct costs are much easier to calculate, they must be interpreted with care, because state payments such as unemployment benefits narrowly represent the costs to the state, while at the same time they correspond to a benefit for the unemployed.  Payments for UI have been implemented precisely to offset the high economic costs associated with unemployment.  Thus they are both a cost and a benefit.

The C-FEPS measure COU-MO (Cost of Unemployment for Missouri) aims to quantify the economic cost of unemployment in Missouri by suitably dealing with the impact of unemployment insurance payments. In our measure, we will strictly focus on the loss of income to persons and the state of Missouri as a whole.  This loss in income includes the personal loss to those who are unemployed but also to all others who would have benefited had this income been spent on goods and services throughout the economy.  In other words, firms lose because of foregone sales and profits, individuals and their dependents lose due to the general fall in income, and the state loses due to foregone sales and income tax revenue.  Our measure also recognizes that the direct cost to the state in terms of UI payments has a mitigating effect on the loss of income from unemployment.

 C-FEPS Measure of the Cost of Unemployment for Missouri (COU-MO)

 The present measure of the cost of unemployment for the state of Missouri has two components.  First, it includes the opportunity cost of foregone income by employable Missouri residents, who are ready, willing and able to work but cannot find employment.  This measure subsumes the forgone (income and sales) tax revenue which the state would have collected had these individuals been gainfully employed.  We look both at the cost to Missouri residents and to the state of Missouri.  However, because the state makes unemployment insurance payments to unemployed individuals, these payments and their corresponding multiplier effects have a dampening and countercyclical effect on the state economy.  Therefore, our COU-MO measure includes a second component that accounts for UI payments, which we compare to what the cost of unemployment would have been had these payments not been made.

TABLE: The Cost of Unemployment in Missouri, 2004

	Not accounting for UI payments	After accounting for UI payments
Total Income Lost	$ 6,736,539,855	$ 5,676,101,577
Total State Income Lost	$ 534,263,187	$ 1,049,439,187
Total Income Lost (per unemployed individual)	$ 37,003	$ 31,178
Total State Income Lost     (per unemployed individual)	$ 2,935	$ 5,764

 

The cost of unemployment in terms of total income lost in the state of Missouri during 2004 was a whopping $5.676 billion.  If the state had not made unemployment insurance payments, it would have been much higher – over $6.7 billion.  Since these figures represent the income lost due to unemployment, they also subsume the lost tax revenue to the state, which amounts to over $534 million.  If we add the direct cost of unemployment insurance to the foregone state tax revenue, we see that the cost to the state rises to over $1.049 billion. 

The table also shows the total income lost to MO’s economy per unemployed individual.  This figure should not be interpreted to mean that, if a jobless individual were employed, he or she would have earned $37,000 in wages.  Rather the figure calculates both the foregone individual wages and their foregone multiplier effects. This is because when an individual is unemployed, she or he would lose not only the earnings but also the benefits that come from buoying economic activity as a result of spending these earnings.  After accounting for UI, the number drops to $31,178 because UI offsets the loss of income by the amount of the UI payment and its multiplier effect. 

Finally the table shows that the cost to the state per unemployed individual is $5,764, which includes the direct cost of paying unemployment insurance and the opportunity cost of foregoing $2,935 in income and sales tax revenue.

Figures 2 and 3 show the historical data for these measures since 1998 (the first year for which adequate state UI data is available). If we look at the trend since 1998 in Figure 2 we see a sharp increase in the cost of unemployment during the latest economic downturn.  Irrespective of which measure we consider (with or without UI payments) the cost of unemployment (unadjusted for inflation) has more than doubled in a matter of six years from its trough in 1999.

 

 

 

 

 

 

 

 

 

 

 

Figure 3 shows a small decline in 2004 in the measure of state costs, which includes unemployment insurance payments.  There is little cause for cheer, however.  In March of 2003 Missouri’s unemployment insurance trust fund went bankrupt forcing the state to borrow money from the U.S. Treasury in order to pay benefits to eligible claimants. This in turn prompted reforms to MO’s unemployment compensation system to restore solvency.  Some of the changes included increases in employer contribution rates and the taxable wage base.  But the reform also established conservative caps on weekly benefits and a series of ever more punitive means test for eligibility.  In sum, in the face of still rising unemployment in Missouri (see Figure 1) and falling state unemployment insurance payments, the safety net which the state provides is being eroded.  Coupled with the aggressive budget cuts on other social spending, the countercyclical effect of state spending will continue to diminish. We should thus expect that, because the state is strapped for cash to relieve the loss in income from unemployment, the cost of unemployment in Missouri would continue to rise in 2005.

 

 

 

 

 

 

 

 

 

 

State Budgets and the Cost of Unemployment

Since Nixon’s Presidency revenue-sharing programs aimed to send federal funds to states for various social programs.  This led to increased state responsibility for program decision-making, and soon thereafter increased responsibility for funding these programs.  The era of devolution resulted in greater burdens on states to provide social services and in fewer federal grants for these programs.  States, however, are not always able to deal with the corresponding financial burden, especially in recessions.  Unlike the Federal government, which can run deficits to fund necessary programs, the states face constitutional and market-imposed requirements to balance their budgets.  (For detailed analysis of the recent state budget crises, see C-FEPS Policy Note 02/05 “The Perfect Fiscal Storm” by L. Randall Wray.) 

As we have seen over the last two years the state of Missouri could not deal with the budget crisis on its own.  To meet UI payments it had to turn to the Federal government to borrow; it also slashed spending and increased taxation.  In April 2005, Missouri implemented aggressive cuts in Medicaid spending – a program which benefits the low-income families and children.  Under the new law nearly 14% of beneficiaries, or over 100,000 parents, and disabled and elderly people, will become ineligible for Medicaid.  Thousands more will have to pay for part of their medical bills, and for others, eligibility requirements will become more stringent. 

What does all of this mean for the state of Missouri?  Quite possibly, it means that the state will be at the end, and not at the beginning, of the economic recovery.  The AFL-CIO reports that since 2001 Missouri has experienced heavy job losses in industries, which generally offer high wages and benefits.  Between January 2001 and October 2004, Missouri lost 33,400 jobs in manufacturing and 9,800 in information technology. Outsourcing too has worsened Missouri’s job picture. One estimate is that 14,677 jobs have moved south of the border as a result of NAFTA (AFL-CIO, October 2004).  This job loss has put downward pressure on wages. Average wages in Missouri’s growing industries are 32% lower than in Missouri’s shrinking industries.  At the same time health coverage steadily disappears from many compensation packages.  In 2003, 620,000 residents of Missouri were uninsured, which is an increase of 18.3 % over just 3 years.  Finally, the AFL-CIO reports that in 2003, 602,000 Missourians were poor, an increase of 16% since 2000, and that personal bankruptcies have increased 46%.

 The Missouri job market has become increasingly more precarious.  Jobs are leaving the state, wages and benefits are falling and more and more people cannot rely on state and federal benefits for basic support.  During the first nine months in 2004 nearly 57,000 people in Missouri exhausted their eligibility for UI before finding employment.  The cost of unemployment in terms of lost income in 2004 was $5.676 billion.  But behind the numbers stands a more fundamental job crisis.

 

---- 

Technical Note

In computing the cost of unemployment we have developed a broader measure for the level of unemployment in the state of Missouri, which contains the officially unemployed as well as those who are discouraged and, therefore, out of the labor force.  Since the Bureau of Labor Statistics does not report statewide figures for discouraged workers, we have extrapolated our numbers from the national data. Discouraged workers should be counted as unemployed because they are ready, willing and able to work, but were not looking for a job in the survey week due to pessimism about the availability of jobs and employment prospects.

We next use median personal income for the state of Missouri to calculate foregone earnings.  Since the unemployed are generally from the lowest income strata, it is reasonable to assume that their propensity to spend out of income is 100% (for some it is in fact greater, as many low-income individuals borrow to spend more than their incomes).  If all income were spent, we would need to account for the multiplier effect of these expenditures.  The Bureau of Economic Analysis reports the spending multiplier for the state of Missouri to be equal to 2.0582.  Thus the total income lost includes both wages and the multiplier effects of spending these wages.

For calculations of the foregone state revenue we use MO’s income tax rate of 6% for incomes over $9,000 and MO’s sales tax rate of 4.225%.  To measure sales tax revenue lost we apply the 4.225% sales tax rate only to 45.7% of income, which is the Missouri income tax base subject to the sales tax (see Bruce and Fox, 2000).

We next take data on Missouri payments of unemployment insurance (UI) since 1998 as reported by the Department of Labor.  These are actual disbursements to residents of Missouri, which reduce the loss of income from unemployment.  To calculate the net loss in annual personal income after UI payments, we also account for the multiplier effect of receiving UI.  For the state, the cost of unemployment is not only the foregone income and sales tax revenue but also the actual direct costs of paying UI to jobless Missouri residents.

Finally we divide the aggregate measures of lost personal annual income by the broader unemployment measure for Missouri to arrive at the foregone income per unemployed individual.  We also perform a similar calculation to get the cost of unemployment to the state per unemployed person.

Why Our Measure is Conservative

We only include the officially unemployed and discouraged workers, but do not account for underemployment or for those individuals who are out of the labor force for reasons other than discouragement, but who want a job.  All of these persons are missed by the official measures of unemployment.  While the underemployed are not technically unemployed, they are also not working in the desired capacity in terms of hours, compensation or skill.  Generally, these are people who work part-time but seek full-time employment or who are in full-time jobs but have worked less than 35 hours in the survey week for economic reasons.  But these also include high-skill workers who take up low-skill jobs that do not utilize their education and experience.

Measuring underemployment is difficult; it is important however to capture these forms of underutilization in the unemployment numbers, because those workers compete for available jobs with those who are officially unemployed. Incorporating all individuals who are underemployed or out of the labor force but would like to get a job would boost considerably our calculation of the cost of unemployment.

Also note that when we calculate the cost to the state we do not account the foregone revenue from loss of collection of local and other taxes that may be imposed in different cities, counties and districts across the state.  However, since the loss of state revenue is subsumed in our overall COU-MO measure, these costs are already implicitly accounted for.  What we do not account for is the increased expenditures of Medicaid and TANF.  Neither are we able to measure the “untold costs” in terms of higher crime, incarceration and divorce rates, as well as health problems and other factors, which we alluded to in the section about indirect costs.

---

REFERENCES:

AFL-CIO. October 2004. “The Missouri Job Crisis.” www.afl-cio.org/yourjobeconomy/jobs/upload/mo.pdf

Bruce, Donald and William J. Fox. December 2000. “E-commerce in the Context of Declining State Sales and Tax Bases.” National Tax Journal. No. 53. Vol. 4. part 3. 1373-1390.

Forstater, Mathew. 2003. “Unemployment”. In J.E. King (ed). The Elgar Companion to Post Keynesian Economics. Chetlenham, UK: Edward Elgar.

Kulp, Clarence A. 1932-33. “Calculations of the Cost of Unemployment Benefits.” Proceedings of the Casualty Actuarial Society. Vol. 19. pp. 268-278.

Okun, Arthur. 1962. “Potential GNP: Its Measurement and Significance.” Proceedings of the Business and Economics Statistics Section of the American Statistical Association. pp. 98-104.

US. Department of Labor, Employment and Training Administration, Unemployment Insurance Program Letter No. 44-81. October 13, 1981. http://workforcesecurity.doleta.gov/dmstree/uipl/uipl81/uipl_4481.htm

Winkelmann, Liliana and Rainer Winkelmann. 1998. “Why are the Unemployed so Unhappy? Evidence from Panel Data.” Economica, v. 65. pp. 1-15.

Wray, L. Randall. 2002. “The Perfect Fiscal Storm”. C-FEPS Policy Note 02/05.