Changes in Consumer Outstanding Credit Debt
College of Business
Western Carolina University
Outstanding credit card debt is forecast to grow approximately 7.13% throughout 1999 and 9.50% during 2000. The forecast is based on the Loanable Funds Model with assumptions that business investment, government budget deficit, and foreign borrowing in the United States are held constant. It was also assumed that interest rates would remain constant. Consumer outstanding credit debt was projected as a e linear function of real personal consumption expenditures lagged two years (24 months) with these same assumptions, using monthly data from January 1992 to December 1998. The expected increase in credit card debt may persuade Americans to pay more attention to their purchases and card balances. Responsibility for controlling increases in debt lies more with consumers than with government or the Fed. Increasing interest rates to discourage credit card usage may do more harm than good. (JEL: E21, E51)
Part 1. Introduction
This paper forecasts changes in consumer outstanding debt for the years 1999 and 2000. The explanatory variable is real personal consumption expenditures. The economic theory used as a basis for the forecast is the loanable funds model. The approach is based on a regression model using the two variables. This approach benefited from simplicity of the formula used to calculate the forecast.

Since 1992, the amount of outstanding credit card debt has significantly increased. It is expected to rise more over the next two years. Inflation has remained fairly stable during the 1990’s. However, it is difficult to determine future changes, which is why a short, two-year time frame is relatively reasonable for forecasting consumer debt.

The rest of this paper is organized as follows: Part 2. presents the data used to forecast consumer outstanding credit debt, which includes real personal consumption expenditures; Part 3. explains the theoretical basis for the approach adopted in forecasting the consumer outstanding debt; Part 4. presents forecasts of consumer outstanding credit debt for 1999 and 2000; Part 5. evaluates the importance of the forecast for the economy; and Part 6. discusses conclusions for economic policy.

Part 2. Data
All variables are taken from the Federal Reserve Bank of St. Louis Federal Reserve Economic Data (FRED). The measures of consumer credit outstanding, and real personal consumption expenditures are FRED variables TOTALSL and PCEC92, which are measured in billions of dollars, seasonally adjusted annual rates. Personal consumption expenditures is a real variable measured in chained 1992 dollars. The consumer price index (CPI) for all urban consumers was used to put consumer credit outstanding in real terms by adjusting for inflation. The CPI is FRED variable CPIAUCSL, with base year 1982-84 = 100, seasonally adjusted. The primary source is the U.S. Department of Commerce Bureau of Economic Analysis.

These variables are measured on a monthly basis. The sample period runs from January 1992 to December 1998. Note the sample period begins near the end of the recession the country was in during the early 1990’s. Real personal consumption expenditures was used as the explanatory variable since a large portion of these expenditures are financed with consumer credit.

The forecast horizon is two years into the future. The regression used is univariate, since real personal consumption expenditures is the only independent variable. The right hand side (RHS) variable is lagged 24 months to provide data for the forecast calculations.

Both sets of data are increasing over time, and appear to be linear. From this, it is assumed that there is a directly proportional relationship between the two variables. This is demonstrated by the graph provided above.

Part 3. Economic Theory
This forecast assumes inflation will remain constant over the two-year time frame. This constant should not have an effect on outstanding credit debt. Since credit cards are used to finance primarily consumption expenditures, this forecast assumes government spending, investment, and net exports will also remain constant.

The theoretical model that is applied to this forecast is the Loanable Funds Model. This model is appropriate since one of the four sources of demand for loanable funds is household credit buying. It is assumed that the other three sources of demand will remain constant. These include business investment, the government budget deficit, and foreign borrowing in the United States. Another assumption is that interest rates remain constant.

The approach for this forecast is that personal consumption expenditures will increase proportionally as outstanding consumer credit debt increases. This model uses two-year (24 month) lagged values for real personal consumption expenditures.

The mathematical model that will be used for the forecast is the univariate model. An example of this model is

Yt = A + B Xt-24

Y represents outstanding consumer credit debt. X represents real personal consumption expenditures, lagged by 24 months. B is the slope and A is the y-intercept of the univariate equation. Slope and intercept coefficients are estimated for the forecast equation by ordinary least-squares regression.

The forecast horizon is two years or 24 months. The model, approach, and assumptions are appropriate for this horizon, since inflation and interest rates are not highly likely to remain constant for more than two years.

Since significant portions of real personal consumption expenditures are financed via credit cards, an increase in personal consumption spending would increase the amount of credit card debt. Using the model and assumptions stated above, as the quantity of real personal consumption expenditures increase, credit card debt would increase and shift the demand curve on the loanable funds model to the right. The supply curve would also shift to the right due to the assumptions of constant interest rates.

One of the shortcomings of this paper’s approach is that interest rates are expected to remain constant over the next two years. However, this assumption is not very certain since the rapid growth of the United States economy may persuade the Federal Reserve to raise interest rates. This would impact borrowing and affect a large portion of consumption expenditures.

Part 4. A Two-year Forecast of United States Consumer Outstanding Credit Debt
The graphs above indicate real personal consumption expenditures and outstanding credit debt are linear, and should be directly proportional. The debt forecasting equation was estimated using an ordinary least squares regression for the period January 1992 to December 1998. The R squared is 0.919. The estimated coefficient for outstanding credit debt is –701.596 with a t-statistic of –12.752. The estimated coefficient for real consumption expenditures is 0.315 with a t-statistic of 25.686. The standard error for the regression was 18.271. The estimated equation used for the basic univariate model is:
Dt = 0.315(25.686)Xt-24 –701.596(-12.752)

A sample calculation for the amount of outstanding credit debt for February 1999 is:

D1999.2 = 0.315(4853.8) – 701.596 = 827.4

According to this calculation, the amount of outstanding credit card debt in February 1999 will be 827.4 billion 1982-84 dollars. The increase to this amount is not necessarily a good thing since many Americans tend to abuse their credit cards. Results from the forecast calculations are provided in Table 1.

Table 1
Forecast Outstanding Credit Debt, Jan-1999 – Dec 2000
(Billions of 1982-84 Dollars, SAAR)
Month Credit debt
Jan-99 826.4
Feb-99 827.4
Mar-99 827.9
Apr-99 828.3
May-99 831.2
Jun-99 840.5
Jul-99 855.4
Aug-99 856.2
Sep-99 858.5
Oct-99 860.6
Nov-99 868.2
Dec-99 873.4
Jan-00 883.2
Feb-00 892.2
Mar-00 896.8
Apr-00 902.1
May-00 915.4
Jun-00 925.7
Jul-00 923.0
Aug-00 929.1
Sep-00 939.9
Oct-00 944.1
Nov-00 946.0
Dec-00 956.4
The forecast growth rate of outstanding credit debt from December 1998 to December 2000 is 20.17%. The two-year growth rate from December 1996 to December 1998 is 7.13%. A comparison of actual and forecast consumer credit is shown in the following graph.
Part 5. Forecast Implications: A Two Way Mirror
This forecast can be viewed favorable or unfavorable, depending upon one’s interest. Outstanding credit debt forecasts are of interest to a variety of people including banks and financial institutions, which are major sources of credit lending. A projected increase in credit card demand would allow card issuers to adjust to the changing needs of its customers. Issuers can adjust interest rates and payment terms to encourage use of their cards.

Retail stores, gasoline retailers, and other businesses that accept credit cards would be interested in this forecast. An increase in credit card debt may indicate higher sales for most of these companies. Many of these companies issue their own credit cards, and may desire to change their credit terms to be competitive with other card issuers.

Consumers may find this forecast unfavorable because the large amount of debt could trigger interest rates to increase, which could hurt them in more ways than one. On the other hand, consumers may benefit from a selection of credit card sources. As mentioned earlier, issuers competing for new customers will be offering better terms and service, benefiting the consumer in the long run.

Part 6. Policy Conclusions
Outstanding consumer credit debt is expected to rise approximately 20.17% from 1998 to 2000.

Assuming credit card debt increases by this amount in the next two years, this may be a good sign for the economy in certain scenarios. If Americans pay credit card balances on time, this would probably indicate an increase in consumption spending. However, if Americans let their balances build up and do not pay them off quickly, most of this increase would be interest on previous outstanding debt. Americans should be aware of the amounts they charge, and how often they use their credit cards.

Credit card balances should be paid on time to avoid unwanted, unproductive finance charges. If interest charges are inevitable, Americans should look for credit cards that offer the lowest annual percentage rate to minimize the total amount spent to cover the purchase. On a personal level poorly managed credit can result in personal bankruptcy, a diminished standard of living, and a battery of other financial and personal problems.

The government is limited in what it can do to regulate credit card debt. Increasing the standards for obtaining credit cards may be one way to deter individuals who do not need a card from applying for one.

Also, increasing the minimum payment on credit cards would help ensure people pay off credit card debt faster.

The Federal Reserve is also limited in what it can do to regulate credit card debt. The main thing the Fed could do is raise interest rates to discourage credit card usage. However, this would do more damage than good, since real estate and other major industries are more sensitive to interest rate changes than individuals with credit cards.