Compromise programming calibration for financial analysis of firms of a common sector of business, case study for a set of Spanish banks in 1995

Abstract

To perform a Financial Analysis (FA) procedure on a set of different firms of the same business sector and region or stock market (maybe with a definite management goal) a set from 7 to 20 balance ratios is chosen as representative of the firms situation, and a benchmark set of good firms of the same kind is selected to calibrate the ratios as attribute variables using Compromise Programming (CP) comparison procedures. For that using the ratios of the benchmark set a reference CP ideal point is obtained, and with it for each ratio a reduced Ratio Quality Index (RQI) is obtained containing a relative CP quality evaluation. From these RQI, a CP Global Utility Index (GUI) of each firm is then proposed, using a common set of chosen CP weights to incorporate the effect of each ratio on the quality of the firms for the intended FA study, with indication of some special techniques including possible use of utility-like functions. An illustrative case study follows with normalized balance sheets data of some main Spanish banks in 1995, showing that some of the real features of this suggestive example are put in evidence by this method, that quality classification depends on various factors, and that external wide information is necessary and gets incorporated with the proposed method. A discussion follows concerning the case, the method and possible future developments.

Acknowledgement

Thanks are given to Prof. Joaquin Millan of the Department of Applied Economics in E.T.S. Ingenieros Agronomos of U.P.M. for the fruitful scientific collaboration and counselling on matters related to this paper.

Notes

¹ The trials with the case study did show that such extension (2)-3 of formula (2) was necessary because without adaptation the (2)-1and (2)-2 alone gave absurd Quality Indexes y_ij  = 1 − z_ij , mostly 0 except 1 if x_ij  = 0, when zo_i (positive by definition) was null, and also when it was near 0.

² Revised in 2004 for ‘simplified accounts regime’, Real Decreto 296/2004, of 20 February.

³ The authors made several other non presented internal case studies with this global set SG of ratios some of them including also the balances of 1994 finding rather similar results, and for that they duplicated the number of ratios before the use of formulae (2) to (11), the ratios of 1994 having weights reduced by a factor of 0.8.

⁴ As known, see ‘www.riskglossary.com/link/basle_committee.htm' from Basel Committee on Banking Supervision to have an excellent official definition linked also to an organization for central banks B.I.S. (Bank for International Settlements) in www.bis.org; these rules demand a capital + reserves cushion roughly as a difference of deposits and other items that can be withdraw by customers to debts to the bank including credits that are assets with specific risks, the rules Basel-1 asking for a more precisely defined 8% cushion, with procedures to make that rule compulsory as an obligation of increasing capital or decreasing risky lent sums, the goal being to get effectively stability for the global banking system.

⁵ They have a probabilistic definition of margin that considers that the risk is lower if the credits of lent sums have independent default chances, and that reduces the exigency of capital or equity for many important banks. The article (Meon, 2005) offers a look about actual financial systems theories.

⁶ The high value for Banesto is explained by the precedent quasi-default; the relatively high values for Hispano because of problems in some subsidiaries (as B. Urquijo because of an older industrial crisis) and for Argentaria because of a low denominator mostly, not having also excellent debtors .

⁷ As known not all sums in deposits are free for lending, regulations impose for macroeconomic control that a proportion r must be kept in cash or equivalent state (sovereign?) deposits, the goal being to limit the ‘multiplicator’ (to less than 1/(1 − r) for extreme geometric growing of deposits that are lended and become deposits, etc …) to avoid uncontrolled growing of monetary masses (sums of money at disposition of economic agents) that could cause a burst of inflation, etc.

⁸ These state credit institutions were impulsion for economic activities during several decades, at first in part to avoid usury for small businesses, and other banks replaced their activity.

⁹ At first dividends and not net earnings, but the part of not paid earnings for reserves intervenes as it is often used for increasing the firm nominal size and future earnings, and that is usual and necessary for a bank, specially if it is growing and stressed for complying with Basel1 or Basel2 rules.

¹⁰ As known an arbitrage is roughly saying a possible operation in markets having positive probability of gains without risk of losses.

¹¹ In the International Finance Congresses (such as the 21st Annual European Finance Association Meeting in Brussels on 25–27 August 1994) there is a mention to ‘presence of some GARCH-effects (Estimation of GARCH Processes with temporal Aggregation: some empirical evidence by D. Dannenburg and B. Jacobsen)’ in stock prices time series. To use time series to earn money speculating is maybe possible but practically not evident, and in theory it would be contrary to efficiency of markets. A wise speculator could earn money building good methods data mining the enormous bulk of data from past stock markets series, on condition that other speculators do not know that because otherwise concurrence would destroy advantage.