Issues in Forecasting Company Liquidity

Article excerpt

In recent years, the rise in corporate bankruptcy has led to an increased interest in the examination of company liquidity. Credit analysts and others involved in the evaluation of a firm's financial position are often concerned with both the measurement of current liquidity, as well as the accurate prediction of future liquidity. Trends in a company's liquidity position make it possible to use past and current information to help predict future liquidity. These measurements are often used to examine the company's financial position and evaluate its ability to meet financial obligations including the likelihood of bankruptcy.

McNamara et. al. (1988), for example, found two different measures of liquidity, the ratios of total liabilities to total assets and of total liabilities to current assets, to be significant predictors of company bankruptcy. Ohlson (1980) also used two measures of liquidity, the ratios of total liabilities to total assets and of current liabilities to current assets, in a bankruptcy prediction model based on accounting data.

In addition to being useful in predicting bankruptcy, liquidity measures also are used in evaluating financing alternatives and the overall credit worthiness of a company. Thus, being able to accurately forecast liquidity would enable loan officers and credit analysts to make better credit and other business decisions.

Random Walk Model Predicts Most Accurately

Past studies have examined the prediction of annual and quarterly earnings, cash flow, and earnings per share. However, few studies have specifically examined the prediction of company liquidity. This study extends previous work in cash flow and earnings forecasts to the area of forecasting annual company liquidity.

Two specific questions are related to the prediction of company liquidity:

(1) Is there a significant difference in forecast accuracy among various models used to predict measures of liquidity, and if so, what is the most accurate forecasting model?

(2) Is there a significant difference in forecast among various liquidity measures, and if so, what liquidity measures are most accurately forecasted?

A significant difference in forecast accuracy exists among different prediction models. Among the six forecast models, a simple random walk model was the best for forecasting each of the eight measures of liquidity examined. Because the random walk model is easy to use and does not require as many observations as other prediction models, it is well suited for short-range forecasts of company liquidity when past information is limited. In addition, the strong performance of the random walk model, relative to the other models examined, indicates that the effort often spent in using more sophisticated forecasting techniques may not be justified in the forecasting of future company liquidity. Loan officers interested in quick methods to forecast a company's liquidity may do just as well with a simple random walk model as with more sophisticated and complex models.

Another finding is that a significant difference in forecast accuracy exists among the different measures of company liquidity. Of the eight short-term and long-term liquidity measures examined, the ratio of total liabilities to total assets was the liquidity measure most accurately forecasted, regardless of the prediction model used. This suggests that some measures of company liquidity are more accurately forecasted than other measures. Those interested in forecasting a company's liquidity should be aware that not all liquidity measures are accurately forecasted on an equal basis.

How to Measure Liquidity

Numerous ratios have been proposed as alternative measures of liquidity. The eight ratios examined are all computed using financial accounting data. Four of these ratios represent measures of short-term liquidity, and four are considered measures of long-term liquidity or solvency. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.