This paper forecasts U.S. real personal savings (RPSAVE) for the years 1999 and 2000. The explanatory variable is the unemployment rate. The approach is based on the new classical macroeconomic theory using the unemployment rate (URATE) from 1992.1 - 1998.4. The new classical macroeconomic approach benefits from simplicity of implementation, and exploits the role played by unemployment in the U.S. in driving real personal savings.

According to our data obtained from the Federal Reserve Bank of St Louis, real personal savings are currently low by historical standards and respond to changes in unemployment. The forecast horizon of two years was chosen to minimize the possibility of external factors impacting the economy in an unforeseen way. The forecast horizon is short enough to avoid seriously overstating or understating future real personal savings.

The rest of the paper is organized as follows: Part 2. presents the data used to forecast real personal savings; Part 3. presents the theoretical basis for the approach adopted in forecasting real personal savings; Part 4. presents forecasts of real personal savings for 1999 and 2000; Part 5. evaluates the importance of the forecast for the economy; and Part 6. discusses conclusions for economic policy.
 

All variables are taken from the Federal Reserve Bank of St Louis Federal Reserve Economic Data (FRED). The measures of nominal personal savings, consumer price index, and unemployment rate are FRED variables PSAVE, CPI, and URATE, and are seasonally adjusted annual rates (SAAR). The nominal variable PSAVE was converted to a real variable (RPSAVE) by doing the following: Real Personal Savings = (PSAVE x 100/CPI). The primary source is the U.S. Department of Commerce Bureau of Economic Analysis. The variables for PSAVE were listed quarterly and for CPI and URATE monthly. The monthly data were transformed into quarterly data by taking the average of the three months that constitute a quarter (e.g., January, February, and March) and using those numbers as the quarterly unemployment rates and CPI. The sample period for the data used is from the first quarter of 1992 to the fourth quarter of 1998. This sample period runs from the end of the 1990-91 recession to the end of the available data. The forecast horizon is two years (eight quarters) into the future. Regression analysis was first done on the original data to establish that a relationship exists between the variables.
 

Assumptions made in this forecast include the following: all other variables are held constant (ceteris paribus) for the next two years and inflation will be held at a constant rate. Hence, there will be no additional impact on personal savings.

Under normal circumstances, there is a short-run trade-off between the inflation and unemployment. As unemployment drops below the natural rate of unemployment, inflation should begin to rise if not held at a constant rate. The quarterly unemployment rates are lagged for eight quarters to project real personal savings for the years 1999 and 2000.

The values for the estimated future unemployment rates are used to calculate the personal savings forecast for each quarter. This forecast assumes real personal savings is a function of the unemployment rate:

This is a simple model appropriate for forecasting short-term personal savings. The forecast projects a proportional change in real personal savings to a change in the unemployment rate and assumes a linear relationship.

Personal savings are forecast to decline over the next two years with inflation remaining at its current rate. As long as this trend of increasingly low unemployment continues, personal savings will continue to decline in order for individuals to maintain their current standard of living. Once personal savings are no longer available or only a small amount remains, acquiring debt is the next alternative to keeping the current standard of living, and spending should slow. With spending slowing down firms will have more inventory (quantity supplied) than customers are willing to purchase (quantity demanded).

With a larger quantity being supplied than being demanded, firms will be forced to lay off employees. This action will drive the unemployment rate up towards the natural rate of unemployment. This response is termed the self-correcting mechanism.

"In theory, the economy exhibits an inherent tendency toward self-correction that should automatically eliminate recessionary and inflationary gaps over time," (Thomas 1996, p. 607). The same holds true for the unemployment rate. When the unemployment rate is either higher or lower than the natural rate of unemployment, then certain factors such as wage rates and inflation, influence unemployment toward its natural rate.

When trying to avoid such a situation new classical macroeconomists believe that discretionary monetary policy is likely to be ineffective and possibly counterproductive. Basically, such policies can destabilize economic activity causing the business cycle to be amplified. New classical economists believe that the ineffectiveness of monetary policy is due to the expectations of the public (Thomas 1996, pp. 608-9).
 

 Equation 2 was estimated with lagged 1992.1-1998.4 quarterly data. The regression estimate is (t-statistics in parentheses):

The estimated a₁ is the coefficient on URATE, indicating that a one percent increase in the unemployment rate will increase the real personal savings rate by 79.565 billion 1992-1994 dollars. The adjusted R-square of the estimate is 0.47, indicating approximately 47% of the variation of RPSAVE is explained by variation in URATE. The t-statistic of a₁ is greater than three, indicating a strong rejection of the null hypothesis that a₁ = 0. The t-statistic of the intercept, a₀, is greater than two, indicating a strong rejection of the null hypothesis that the intercept equals zero.

Forecast real personal saving calculated from the regression estimate is presented in Table 1.
 

Table 1 Forecast Real Personal Savings in Relation to the Unemployment Rate, 1999.1 – 2000.4 (billions of 1992-1994 dollars)
Quarter	Percent Change in RPSAVE	Projected Change in RPSAVE	Projected RPSAVE
1999.1	3433.78%	101.64	98.68
1999.2	-19.91%	-19.65	79.03
1999.3	-6.69%	-5.29	73.74
1999.4	-23.57%	-17.38	56.36
2000.1	-5.36%	-3.02	53.34
2000.2	-32.58%	-17.38	35.96
2000.3	21.00%	7.55	43.51
2000.4	-17.35%	-7.55	35.96

 
This forecast projects real personal savings to decrease approximately 43% from 1999.1 to 1999.4 and 33% from 2000.1 to 2000.4. This represents a steady decrease in the amount of real personal savings with only one explanatory variable, the unemployment rate. These are reasonable projections as long as the unemployment rate continues to decline and there are no real changes in wages or inflation.
 

Real personal savings forecasts are interesting to a broad range of people because personal savings are affected by economic activity. Because the forecast presented here can be interpreted as a decline in personal savings in relation to an increasingly low unemployment rate, it could be compared to the U.S. Department of Commerce’s projected personal savings data to provide information on consumer spending. The forecast predicts a proportional decrease in real personal savings in relation to the unemployment rate.

The forecast suggests that as unemployment declines personal savings will decrease also, and vice versa. Over the next two years the unemployment rate is expected to remain low, indicating industries are retaining their workers and adding more jobs. With this trend individuals believe their jobs are stable and will continue into the future. Therefore, people are spending more than earned causing a decrease in real personal savings.
 

The unemployment rate is declining at approximately 0.5% per year (e.g. 1988.1 URATE at 5.2% and 1989.1 URATE at 4.7%), with real personal savings decreasing approximately 64%, from 1999.1-2000.4, over the next two years.

Assuming the forecast turns out to be correct, individuals are confident that they are able retain jobs and counting on a stable economy. With this assumption, they are consuming more and therefore reducing their personal savings and increasing their standard of living. This increase in the standard of living is able to continue until personal savings are depleted or close to being depleted. When personal savings are gone and acquiring debt is necessary to continue the current elevated standard of living spending should slow. With slowed consumer spending, firms may find it necessary to lay off workers. This response to slowed spending will push the rate of unemployment up toward the natural rate.

Assuming a correct forecast, manufacturing and service firms should plan on a slowdown in the economy. Manufacturing firms could consider keeping their inventories slightly below demand and raise their prices to whatever is determined that the economy can handle. This will increase their profit margins during this increase in spending, allowing firms to avoid acquiring a great deal of inventory that could become difficult to liquidate. By increasing profit margins during increased spending, firms will be better able to handle a recession.

According to the new classical macroeconomics, government intervention would be either ineffective or counterproductive. Therefore, the government and the Federal Reserve should allow the economy to correct itself back toward the natural rate of unemployment.
 

Federal Reserve Bank of St. Louis, Federal Reserve Economic Database (FRED), http://www.stls.frb.org/fred/

Thomas, Lloyd B. Money, Banking, and Financial Markets, New York NY: McGraw-Hill Company, 1996.