On the long-term stability and cross-country invariance of financial ratio patterns

The purpose of this study is to develop empirically-based classification patterns for twelve commonly used financial ratios and to measure the long-term stability and structural invariance of these patterns. The data are based on annual reports of U.S. and Finnish industrial firms for the periods 1947-1975 and 1974-1984, respectively. The selected financial ratios are, according to a priori classification, the measures of short-term solvency, long term-solvency, profitability and efficiency. Classification patterns of the financial ratios are developed using factor analysis and the stability and invariance analyses are carried out using transformation analysis. The results show that empirically-based classifications are not fully equivalent to the a priori classification. The following factors are found: solvency, profitability, efficiency and dynamic liquidity. The empirical results are based both on the value- and equal-weighted indices of the ratios. Classification patterns are developed using ratio indices in the first-difference form. The use of the first differences becomes necessary because of the clear trend in the time series. Further, empirical results show that different aggregation methods lead to different results. The theoretically better value-weighted indices (in the first-difference form) give more accurate and easy-to-interpret empirical results. Factor patterns based on these indices display also very clear time-series stability and cross-sectional invariance. This result confirms the great importance of aggregation method in ratio analysis.