Does MVA measure up? It may be popular and lauded by many, but market value added, argue Stephen Keef and Melvin Roush, does not necessarily give an accurate picture of wealth creation. (Feature MVA).
Many agree. Much praise has been heaped on MVA and its related measure, economic value added (EVA), and many companies have reported great financial success after adopting the techniques. Fortune magazine has published the rankings of companies based on MVA performance since 1993.
Stewart defines MVA as "the difference between a company's fair market value, as reflected primarily in its stock price, and the economic book value of capital employed". Economic book value is the best estimate of the monies shareholders have invested in the company. It consists of the sum of the monies subscribed for new shares, together with retained earnings, which are profits that could have been paid out as dividends, but have been reinvested for shareholders.
The classic approach to wealth generation has been to invest in projects with a positive net present value. One has a view of the future that can be characterised as a time series of expected cash flows. These are discounted to their net present value using the appropriate risk adjusted discount rate. The cash flows could arise from: a new project; an increase in the productivity of an existing project; or from the sale of surplus, or underperforming, assets.
New York consultancy Stern Stewart & Co has used these to market EVA and MVA. Its ideas are not new, but it has successfully differentiated its products from the old approach.
For wealth maximisation, it recommends applying generally accepted accounting procedures to the view of the future. The resulting profits, in conjunction with the asset values in the balance sheet, are then converted into a series of expected EVAs, which are discounted to a present value. Rather than urge the maximisation of net present value, it stresses that one should maximise MVA and EVA. While it is understandable for Stern Stewart to champion its own performance metric, some may see it in the same light as driving from London to Bristol via Newcastle.
Stern Stewart's approach to EVA and MVA has been canny: it presents them as being closer to economics than to accounting. It believes that divisional managers should be able to understand the mechanics of EVA and MVA without being experts in accountancy.
Part of MVA's popularity may be its apparent ability to convert accounting numbers into useful measures of economic value. One would be forgiven for thinking that the holy grail of accountants is to establish the value and relevance of figures, particularly profit. A big breakthrough in this search was made by R Ball and P Brown (2) more than 30 years ago. They studied risk-adjusted share price reaction to the announcement of unexpected profit and found that it accounted for around 10 per cent of the abnormal share price reaction.
MVA won ringing endorsements from the CEOs of leading companies: "It is the way to keep the score. Why everyone doesn't use it is a mystery to me," (Roberto Goisoeta of Coca Cola); "MVA gauges true economic performance," (Art Krause, CFO of Sprint). MVA seemed to be the ideal measure.
Of course, it is not as simple as that. One of the main uses of performance measurement is to provide information about the results of effort expended in the recent past. The trouble is that MVA is a hybrid statistic. Is it an ex-post or an ex-ante performance measure?
The equity book value number in the MVA equation represents investments made in the past, so it is clearly an ex-post measure. Market value is the present value of the future cash flows, which makes it an ex-ante statistic. The difference between those two figures, the MVA, is neither ex-ante nor ex-post. Therefore it is not clear what MVA measures.
MVA is the net present value created for shareholders over the life of the company. It therefore follows that MVA is the present value of future EVA. Market value consists of two components: the value created (MVA or net present value) and the cost of the investment.
There are two things wrong here. First there is no systematic link between the cost of the investment and resulting net present value. It is easy to think of examples of small investments that created large gains (Microsoft, Dell Computer) and examples of large investments that incurred heavy losses (General Motors and Ford). So the value created is no more or less important than the cost of the investment. Individually, they are meaningless.
Second, the only measure of importance is their sum, that is, market value. The net present value created or destroyed last month, or last year, is unimportant. It is the same for economic book value. In the jargon of management accountants, both are sunk costs. What matters is current wealth and how it will change over the next period.
What about predicting the financial future? Consider a firm that has experienced big wealth increases in the past. In the future, the expected wealth increase, through change in share price, is just the standard risk-adjusted expected rate of return. Indeed, two firms with the same beta would have the same expected return even though one has a large MVA and the other a large negative MVA.
This is because all relevant information is fully factored into the current share price if the market is semi-strong form efficient. IBM is a good example. In 1985 it was top of the MVA rankings. By 1995 it was ranked 997, but had climbed back up to 16th position by 1998. In contrast, Microsoft rocketed to the top in just a few years. Unless these are the exceptions that prove the point, MVA is of limited use as a predictor of the future. Share prices are far more volatile than economic book values. The implication is that strategic decisions affect share prices--assets employed by the firm are of far less significance.
Size is a problem for MVA. It is well recognised that, all other things being equal, larger firms tend to produce larger profits. One answer to this is to standardise the reported profit to a rate of return statistic--for example, return on assets employed or return on equity. MVA standardised by an appropriate measure of size would be a more suitable statistic, something Stewart's book acknowledges.
Dell Computer, for example, was ranked 42 in Fortune's 1998 list of America's greatest wealth creators. Its MVA was a staggering $25.7 billion on an asset base of just $0.5 billion. Its MVA to economic book value ratio was more than 50. By ignoring the size effect, Dell's ranking has been grossly understated.
Periodic change in MVA can be used as a performance measure. The change is calculated by the change in market price less the change in book value. The change in book value can be seen as the new equity contributed by shareholders. Therefore, the change in MVA represents the net present value created for shareholders during the period in question.
Changes in market value occur almost continuously. Share prices are routinely reported but, because of the timing of financial reports, a quarterly change in MVA is usually the best that can be achieved. So is it the MVA since inception or the recent change in MVA that is of most significance to shareholders? The answer should be clean A ranking of companies by MVA created in the past year would make interesting reading.
For any interval shorter than the ordinary reporting period, the change in MVA is simply the change in market value. Why bother calculating changes in MVA when an inspection of share price changes will reveal the same picture? This differential measure is applicable only for a short time: it does not take into account the risk-adjusted rate of return that is expected in a competitive market.
The MVA of the firm represents the MVA of shareholders who bought their shares when the firm was set up. But what is the MVA of recent investors? Theirs is the change in the market price of the share since the purchase. Since shares are always changing hands, there are almost as many MVAs as shareholders.
The annual turnover on a competitive stock market is about a third of market capitalisation. A conservative assumption that the half-life of a share is two years--in other words, half the shares in the average company are sold every two years. After 10 years, only 3 per cent of shareholders will retain their shares. So it could be argued that each shareholder has a different MVA. What about shareholders who bought half of their shares a year ago and the other half last week?
Having examined the MVA created by CEOs in relation to the capital available at the time they assumed their position, Al Erhbar, a senior vice-president of Stern Stewart (3), tacitly acknowledges that MVA depends on when shareholders purchase their shares. The implication is startling; there is no such thing as an MVA that is applicable to all shareholders.
By promoting MVA, Stern Stewart has focused on the true goal of CEOs--to create wealth for shareholders. But there are drawbacks. Rather than maximise MVA, CEOs should seek to maximise the orthodox risk-adjusted abnormal return. This is what interests shareholders and measures wealth creation in a competitive capital market.
(1.) G Stewart, The Quest for Value, Harper Business, New York 1991
(2.) R Ball and P Brown, "An empirical evaluation of accounting income numbers", Journal of Accounting Research, Vol 6,1968
(3.) A Ehrbar, "Wealth-creating CEOs: 1st annual MVA rankings", Chief Executive, No 143,1999 A Erhbar, "Using EVA to measure performance and assess strategy", Strategy and Leadership, Vol 27 No 3 1999
Stephen Keef and Melvin Roush are senior lecturers at the faculty of commerce and administration at Victoria University of Wellington, New Zealand