We Need to Measure, Not Count

by Peter F. Drucker

Quantification has been the rage in business and economics these past fifty years.  Accountants have proliferated as fast as lawyers.  Yet we do not have the measurements we need.

            Neither our concepts nor our tools are adequate for the control of operations, or for managerial control.  And, so far, there are neither the concepts nor tools for business control – i.e., for economic decision making.  In the past few years, however, we have become increasingly aware of the need for such measurements.  And in one area, the operational control of manufacturing, the needed work has actually been done.

‘Activity-Based’ Accounting

            Traditional cost accounting in manufacturing – now 75 years old – does not record the cost of nonproducing, such as the cost of faulty quality, or of a machine being out of order, or of needed parts not being on hand.  Yet these unrecorded and uncontrolled costs in some plants run as high as the costs that traditional accounting does record.  By contrast, a new method of cost accounting developed in the past 10 years – called “activity-based” accounting – records all costs.  And it relates them, as traditional cost accounting cannot, to value added.  Within the next 10 years it should be in general use.  And then we have operational control in manufacturing.

            But this control will be in manufacturing only.  We still will not have cost control in services: schools, banks, government agencies, hospitals, hotels, retail stores, research labs, architectural firms and so on.  We know how much a service takes in, how much it spends and on what.  But we do not know how the spending relates to the work the service organization does and to its results – one of the reasons the costs of hospitals, colleges and the post office are out of control.  Yet in every developed country, two-thirds to three-fourths of total output, employment and costs are in services.

            A few big banks are just beginning to implement cost accounting for services. Though results so far are quite spotty, we have found out a few important things: in contrast to cost accounting in manufacturing, cost accounting for services will have to be top-down, starting with the cost of the entire system over a given period. How the work is organized matters far more than it does in manufacturing. Quality and productivity are as important to cost in services as is quantity of output. In most services, teams are the cost center rather than individuals or machines. And in services, the key is not “cost” but “cost effectiveness.” But these are still only beginnings.

            Even if we had the measurements we need for manufacturing and for services, we would still not have the true operational control. We would still treat the individual organization – the manufacturer, the bank, the hospital – as the cost center. But the costs that matter are the costs of the entire economic process in which the individual manufacturer, bank or hospital is only a link in the chain. The costs of the entire process are what the ultimate customer (or the taxpayer) pays and what determines whether a product, a service, an industry or an economy is competitive. A large part of these costs are “interstitial” – incurred between, say, the parts supplier and manufacturer, or between the manufacturer and distributor, and recorded by neither.

            The cost advantage of the Japanese derives in considerable measure from their control of these costs within a keiretsu, the “family” of suppliers and distributors clustered around a manufacturer. Treating the keiretsu as one cost stream led, for instance, to “just-in-time” parts delivery. It also enabled the keiretsu to shift operations to where they are most cost effective.

            Process-costing from the machine in the supplier’s plant to the checkout counter in the store also underlies the phenomenal rise of Wal-Mart. It resulted in the elimination of a whole slew of warehouses and of reams of paperwork, which slashed costs by a third. But process-costing requires a redesign of relationships and changes in habits and behavior. It requires compatible accounting systems where organizations now pride themselves on having their own unique method. It requires choosing what is cost effective rather what is cheapest. It requires joint decisions within the entire chain as to who does what.

            Similarly drastic are the changes needed for effective managerial control. Balance sheets were designed to show what a business would be worth if it was liquidated today. Budgets are meant to ensure that money is spent only where authorized. What managements need, however, are balance sheets that relate the enterprise’s current condition to its future wealth-producing capacity, both short-term and long-term. Managements need budgets that relate proposed expenditures to future results but also provide follow-up information that shows if promised results have actually been achieved.

            So far, we have only bits and pieces: the cash-flow forecast, for example, or the analysis of proposed capital investments. Now, however, for the first time, some large multinational companies – American and European – are beginning to put these pieces together into “going-concern” balance sheets and “going-concern” budgets.

            But most needed – and totally lacking – are measurements to give us business control. Financial accounting, balance sheets, profit-and-loss statements, allocation of costs, etc. are an X-ray of the enterprise’s skeleton. But much as the diseases we most commonly die from – heart disease, cancer, Parkinson’s – do not show up in a skeletal X-ray, a loss of market standing or a failure to innovate do not register in the accountant’s figures until the damage has been done.

Only Beginnings

            We need new measurements – call them a “business audit” – to give us effective business control. We need measurements for a company or industry that are akin to the “leading indicators” and “lagging indicators” that economists have developed during the past half-century to predict the direction in which the economy is likely to move and for how long. For the first time big institutional investor, including some of the very large pension funds, are working on such concepts and tools to measure the business performance of the companies in which they invest.

            These are only beginnings. And so far each of these areas is being worked on separately. Indeed, the people working in one field – for example, pension funds – may not even be aware of the work done in other areas.

                It may take many years, decades perhaps, until we have the measurements we need in all these areas. But at least we know now that we need new measurements and what they have to be. Slowly, and still groping, we are moving from counting to measuring.  (This article appeared in the Wall Street Journal, April 13, 1993)                                          

Mr. Drucker was professor of social sciences and management at the Claremont Graduate School in California until he died in 2005.