College of Business
Western Carolina University
Abstract
South Carolina’s unemployment rate is expected to drop about 1½ percent over the next two years. This forecast is based on lagged unemployment and the index of leading indicators from the years 1980 until 1999. These two variables were used because of the strong correlation between them. The results of a regression of these two variables revealed that unemployment will continue its current trend of decreasing. Since the government is in favor of low unemployment, they will most probably continue efforts to keep it this way (JEL: E24, E66).
Part 1. Introduction
This
paper forecasts the monthly unemployment rate for the years 2000 and 2001. The
explanatory variables are lagged unemployment and the index of leading
indicators. The sample period for this data is from January of 1980 until
December of 1999.
When
unemployment is low, employers will usually increase wages in an attempt to get
more help. However, when unemployment is high, wages will slowly decrease, but
workers don’t want to work for less than normal. This causes unemployment to
linger. It is important to predict the unemployment rate so employers can
anticipate this. Forecasting unemployment is also helpful to people planning to
enter the workforce in North Carolina as well as South Carolina. The forecast
of unemployment is for only two years because the further into the future one
tries to forecast the more inaccurate the results become.
The rest of this paper is organized as follows: Part 2. presents the data used to forecast unemployment for 2000 and 2001; Part 3. presents the theoretical basis for the approach adopted in forecasting unemployment; Part 4. presents a forecast of unemployment for 2000 and 2001; Part 5. evaluates the importance of the forecast for the economy; and Part 6. discusses conclusions for economic policy.
Part 2. Data
The
data used are monthly lagged unemployment and the index of leading indicators
from January of 1980 until December of 1999. The measure of South Carolina
unemployment was taken from the Bureau of Labor Statistics, variable
LAUST45000003. Seasonally adjusted unemployment could have been used for a more
detailed forecast but the Bureau only reported it at a federal level, not a
state level. Consequently, this unemployment rate is not seasonally adjusted.
The index of leading indicators, variable LEAD, was found in the Federal
Reserve Bank of St. Louis Federal Reserve Economic Data (FRED) and it is
seasonally adjusted. The forecast horizon is two years into the future.
One
could also use inflation, or exports to obtain a prediction of unemployment.
This prediction may or may not be predictable, however. This is never known
until executed.
The
index of leading indicators and past unemployment can be used to predict future
unemployment due to trends. The regression analysis, which was conducted on
these data sets, supports this. No transformations other than lagging were
performed on the data.
Part 3. Economic Theory:
Lagged
monthly unemployment rates are used to project future unemployment in South
Carolina for the years 2000 and 2001. The index of leading indicators was
lagged to project the same thing. This yields the forecasting equation, which
can be used to predict unemployment two years into the future. This forecast
assumes unemployment is a function of lagged unemployment and lagged
indexes.
ut
= f(ut-2, indt-2)
Part 4. Empirical Results: Forecast of South Carolina Unemployment
Rates for 2000-2001
The
unemployment forecasting equation was estimated using 1980-1999 monthly data.
The simple regression method was used to find the following results.
Regression Estimate of
Unemployment Rate Forecasting Equation:
Explanatory Variable |
Estimated Coefficient |
T-statistic |
Intercept |
49.42287318 |
17.02725632 |
X Variable 1 |
-0.227840758 |
-3.72961877 |
X Variable 2 |
-0.426599491 |
-16.01697062 |
In
addition, from 216 observations, R square turned out to be 0.63, which is
pretty good. Also, the adjusted R square came out to 0.626. This indicates that
approximately 63 percent of the variation in unemployment can be explained by
the explanatory variables.
The
following table reports the forecast of unemployment rates in South Carolina
for all months of the years 2000 and 2001:
Year Month Forecast
2000 |
Month 1 |
3.758262 |
2000 |
Month 2 |
3.655974 |
2000 |
Month 3 |
3.77571 |
2000 |
Month 4 |
3.752926 |
2000 |
Month 5 |
3.684574 |
2000 |
Month 6 |
3.610406 |
2000 |
Month 7 |
3.303062 |
2000 |
Month 8 |
3.416982 |
2000 |
Month 9 |
3.485334 |
2000 |
Month 10 |
3.397106 |
2000 |
Month 11 |
3.297726 |
2000 |
Month 12 |
3.257974 |
2001 |
Month 1 |
2.885186 |
2001 |
Month 2 |
2.731514 |
2001 |
Month 3 |
2.939478 |
2001 |
Month 4 |
2.913786 |
2001 |
Month 5 |
2.740238 |
2001 |
Month 6 |
2.498338 |
2001 |
Month 7 |
2.393142 |
2001 |
Month 8 |
2.43871 |
2001 |
Month 9 |
2.413018 |
2001 |
Month 10 |
2.32479 |
2001 |
Month 11 |
2.242378 |
2001 |
Month 12 |
2.162874 |
This
forecast is favorable and it accordingly reflects the booming economy.
Part 5: Forecasting
Implications: A Continuing Decrease in
Unemployment
South
Carolina has one of the higher unemployment rates of the states. However, this
hopeful forcast predicts that the unemployment level will continue its current
trend of decreasing. Current markets for employment should become increasingly
stronger and people looking for employment should have less trouble finding
it.
In theory, when the unemployment level decreases, the government pays less unemployment benefits. Therefore, the government, the Fed and most firms would tend to want to continue to influence unemployment rates to stay low. The government should not participate in any activities that would raise it back up again.
Part 6. Policy Conclusions
It is expected that the unemployment rates in South Carolina will continue their current trend and fall into the year 2002. In December of 1999, the unemployment rate was 3.9 percent. The projections gathered estimate that this rate will fall about ¾ percent per year to 2.16 percent by December of 2001.
References
Federal
Reserve Bank of St. Louis, Federal Reserve Economic Data (FRED) (12-30-99), http://www.stls.frb.org/fred/
(1-29-00)
Bureau
of Labor Statistics, Bureau of Labor Statistics Data, http://146.142.4.24/cgi-bin/surveymost