Managing Resource Allocation: Personal Reflections From a Managerial Perspective

In economics, the problem of resource allocation is central. In effect, macroeconomics describes how the allocation of resources affects the behavior of a nation’s economy. Microeconomics, sometimes called the theory of the firm, is supposed to describe how a firm allocates resources and what the consequences will be. When I was a graduate student in economics in the early 1960s, I was assured that macroeconomics was a robust theory, but microeconomics was a very problematic field. At the time, I had no reason to consider seriously either of these propositions. But I knew that how firms allocated resources was important, so I went on to study that process.

In management, the problem of resource allocation is the essence of strategy. Using Mintzberg’s (1978) language, all the words on paper and analytics are empty intention unless scientists, engineers, planners, marketers, sales people, and dollars—the resources of the firm—are allocated to attempt to make that intention a reality. Realized strategy, what the firm actually does, is the revealed consequence of a pattern of resource allocation (Mintzberg, 1978). This in fact is how Chandler (1962) discusses strategy in his seminal work, Strategy and Structure. It is the idea captured in the title of my book with Clark Gilbert, From Resource Allocation to Strategy, which summarizes 35 years of research following my initial study of resource allocation (Bower & Gilbert, 2005).

From the perspective of more than 50 years, what is incredible to me is that very little has changed in how resource allocation is discussed. Despite a great deal of work that describes how the process of resource allocation actually works in firms, the established academic canon has, if anything, become more distant from what actually happens. I should have expected nothing different. Research in the fields of economics and financial economics takes the firm and its shares to be the unit of analysis. Research that examines how firms actually work takes place a level of aggregation below those fields.

My own work Managing the Resource Allocation Process (Bower, 1970), conducted in the 1960s and published early in the 1970s, showed that in the large organizations that characterized the economy, the resource allocation process was dispersed across the business divisions and several levels of the hierarchical organizations used to manage companies. The behavior of the executives in these corporations is governed by the organization design, the planning and budgeting systems, the measurement and information systems, and the incentive systems used to manage activity. The framing of business plans and capital budget proposals reflect what executives believe they are being asked to achieve. Rather than the words on paper that top executives issue in strategic plans or published in annual reports, operating managers try to succeed in the jobs they have been given and for which they are paid, where, as is often the case, they are paid for meeting short-term financial objectives, that is, how they use the resources available to them. If instead they are rewarded for ambitious plans for growth, they seek funding for projects that will generate fast growth if implementation is successful (Bower, 1970; Bower & Gilbert, 2007). The use of the present tense in this summary is intentional since subsequent research has revealed this basic description to be accurate (Bower & Gilbert, 2005).

After Managing the Resource Allocation Process was published in 1970, the review in the Journal of Finance said that the work was interesting but did not seem to understand that resources were allocated the way capital budgeting theory said they would be allocated. It took me a long time to understand that when microeconomists were confronted with evidence that the behavior about which they theorized did not conform to the accepted model, the problem was not the model but that managers did not know how to behave.

Fortunately, management practice was often able to develop independent of microeconomic theory. The management consultants were particularly useful in this respect. The Boston Consulting Group (BCG) made two important contributions. In the 1960s, its work on the experience curve demonstrated that long-run cost curves might continually decline in a fixed relationship to cumulative unit production (Henderson, 1968). This work was of fundamental importance since capital budgeting theory relies on the idea that long-run cost curves rise and therefore the marginal return on a capital investment will decline at some point. This same body of work by BCG pointed out that investment in cost reduction projects in a hypercompetitive industry might have positive accounting returns for a business in that industry, but unless it fundamentally changed the competitive position of the business, it would not change the prospects for negative returns in that business.

The idea that the investment in a project was somehow dependent on the success of the business in which the project was imbedded was not contradictory to conventional capital budgeting; it was simply a largely ignored idea. It was assumed that the “business case” for a project would include appropriate analysis. One way to understand what had happened is that the tools of marginal financial analysis used to examine which of two machines might be the best investment to do a job, or whether a machine should be purchased or rented, were regularly applied to the question of whether a company should invest in a particular business. Should a company enter a business or should a company invest in a strategic expansion of a business were treated in exactly the same fashion as a make-or-buy decision for a machine.

It may not be obvious why these decisions are so different and pose such a different question for resource allocation in the firm or why the difference is so important. The first decision—let us say it is which of two machines a business should buy—is predicated on a number of assumptions that are reasonable to make. To begin, the same machine will be used in the same factory of the same business by the same people. The prospects of the business may be uncertain, but that may affect the usefulness of the two machines more or less equally. Technology is not at issue. Nor is demand. So it is merely a question of cost and availability of capital measured against the likely return. The analytic work involved is well within the purview of department or divisional operating executives responsible for using the equipment.

Now consider the question of whether to enter a business. The attractiveness of the idea depends on a number of future developments that will affect demand. Market structure is at issue, especially the impact of new competitors and new technology. Whether to invest in new technology in the same industry is an equally problematic choice. The way to assess these decisions involves careful strategic analysis. The knowledge and understanding necessary to make those assessments is distributed across the levels and functional lines of the company. But traditional capital budgeting systems used to assess such proposals against the use of capital in other needs make it easy to ignore the key questions that should be considered in making of a strategic decision. Indeed, the focus in most capital budgeting systems is on quantitative summary ratios, such as return on investment, that when compared with the weighted average cost of capital will in principle (in microeconomic theory) indicate which proposals should be funded.

In fact, the range of uncertainties involved mean that it is typical for top managements to look to the track record of the managers who endorse the proposals in order to determine which projects to fund. In effect, with many questions to consider in a limited amount of time, there is a kind of triage. While following formal processes, they review some proposals in a cursory fashion and effectively delegate some of their decisions to the line managers that they trust. Again, in a typical situation, since the list of acceptable proposals exceeds funds available, funds are allocated in some version of a game of fair division in which the principal line executives all receive some funding for their projects. The critical factor here is that denying executives funding for their projects is effectively denying them the chance to grow their businesses and thereby achieve the kind of record that would permit them to progress further in the company.

Where the attraction of a new business or new technology is great, top managements often turn to consultants for advice. When there is considerable uncertainty and executives in the company have limited experience, top management may turn to acquisition as a way to buy the knowledge and management skills their own organization lack. Research suggests that acquisition as a route to entry in a new line of business brings its own special risks and difficulties.

The forces described here are very powerful. Involved, for example, is the question of what it means to give executives a job and then deny them the resources to do it. Obviously if adequate resources are not available, one should not stay in a business. But where resources are being allocated to projects, it is easy to spread them too widely. The essence of strategy is focus. But if focus means denying resources to executives one admires and has high hopes for, it is easy to diffuse resources instead. It is human to fudge such choices.

The documenting of experience curves that showed the existence of more or less perpetually declining long-term costs posed an important conceptual challenge for resource allocation in the firm, further complicating the challenge for top managements. Possession of the low cost position in a market is a huge competitive advantage. But if total unit costs decline with cumulative long-term volume, as shown in a wide range of experience curves, then a firm has to keep investing to protect its position against an entrant willing to invest to gain experience. The battle is unending. If a firm, say, a chemical company, enjoys strong positions in several markets—for example, polyethylene resin and film, polypropylene film and fiber, vinyl plastics, and epoxies—how can it fund the investments needed to sustain position in all of those businesses? And if it cannot, how can it choose which businesses to keep? A more dramatic example might be GE, which in the 1960s had strong positions in computers, nuclear power, and engines for supersonic aircraft as well as electric power generation, distribution, and use and emerging businesses in health care. Spreading resources to meet all requests was impossible. But underfunding a business was suicide.

What was needed was a framework for comparing the strategic position of businesses as opposed to projects that would relate to their long-term profitability. At GE, strategic planners began working on this problem in the 1960s. Using the history of GE’s hundreds of individual businesses as data, they demonstrated the proposition that market share was an important predictor of profitability. Indeed, they concluded that only the businesses with a number 1 or 2 position in a market would earn profits above their cost of capital. This group of analysts spun out of GE to form the Marketing Science Institute (MSI). MSI went on to further elaborate the relationship of market position to profitability. Back at GE, these ideas provided part of the foundation for Jack Welch’s restructuring of GE in the 1980s. Resources were focused on a limited number of businesses. The rest were sold.

BCG’s second contribution to the management of the resource allocation process was related to the same empirical observation (Henderson, 1970). It proposed an industry growth–market share matrix that displayed the businesses of a company according to the growth rate of the industry in which it was participating and its relative share of its market. Relative share was a refinement of the number-1-or-2 idea. High relative share implied that one had more cumulative experience and therefore lower total unit costs. Being number 1 with a dominant share was strategically stronger than a number 1 with a small margin over number 2 and number 3.

Using the matrix, businesses could be characterized according to their strategic characteristics. A business with a high relative share in a rapidly growing business would be profitable and need capital. A business with high relative share in a slow-growth business would be profitable but throw off cash. Low relative share in a high-growth business was a risky but potentially high-return proposition requiring judicious funding. And unless something very dramatic was done to change the circumstances, low relative share in a low-growth business meant continual losses. Sale or exit was justified.

Several of the other major consulting groups developed their own matrices for categorizing and displaying the prospects of businesses. The importance to the management of businesses should not be underestimated. Instead of resource allocation focused on the cost and return of individual projects, managers were now asked to consider the comparative strength of businesses in their portfolio and make strategic choices rather than financial ones. In effect, thinking about a portfolio of businesses enabled firms to break out of a resource allocation paradigm that was effectively a mechanistic rationing of capital according to forecasts of project returns.

This was a conceptual breakthrough. As shown by my research on resource allocation and that of subsequent researchers, firms continued to invest in high-return cost reduction projects in slow-growth, hypercompetitive industries only to find that the businesses continued to lose money even where the calculated return on the project might be high (see, for example, Sull, 2005). A portfolio analysis of a company’s businesses would show that some businesses had to be divested, if resources were to be made available to the strong and promising others.

An associated breakthrough was investment in executive education for subunit general managers. The critical point was that the resource allocation process required as a critical first step the submission of strategic business plans. These were the crucial input to the matrix analysis. And of course, that was the next major challenge to be faced.

Just as project proposals reflected the relative skills and ambitions of the functional managers who prepared them and the general managers who endorsed them, the strategic plans were no better than the ability of those same managers to think strategically. A whole industry of strategic consultants developed to aid middle managers in preparation of plans.

The stakes were high. A plan that resulted in a business being classified as low growth/low relative share was a ticket to business sale. That prospect might be a correct conclusion from the perspective of corporate resource allocation, but it was not one that unit managers flocked to embrace. So companies developed large strategic staffs to “scrub” proposals for the accuracy of their analysis.

While intellectually the framing of the resource allocation process had improved, the resource allocation was still largely a bottom-up affair that depended heavily on the quality of the subunit planning. Because the stakes were so high for the individual managers involved, if the bureaucracy of the company was allowed to run the process, it could readily be corrupted in a way similar to the project-based system it was designed to replace.

What was desperately needed was an honest and open system of resource allocation that reflected the realities of an increasingly complex and competitive global business environment. The 1980s and 1990s saw a series of individual corporate restructurings that served as models for that process. Perhaps most famous was Jack Welch’s restructuring of GE. He dramatically pruned GE’s portfolio of businesses, for example, trading the businesses of GE’s RCA subsidiary for the health care business of the French company Thompson. ¹ As well, he laid off half of the middle management structure of GE—some 200,000 managers. What was once a multilevel hierarchy of more than 100 strategic business units was now a flat organization of 15 companies reporting to a much-reduced corporate office. Business strategies were agreed upon in cross-level meetings that included operating and probable capital budgets. Emphasis was on speed and quality.

With less fanfare, other companies made similar shifts, emerging as globally competitive players. IBM, for example, went through two transformations. The first under Lou Gerstner changed five regional companies into one IBM with five lines of business. Later, Sam Palmisano changed that company into a global provider of digital platforms for the mature and emerging markets. Ginni Rometty, his successor, is now working to transform IBM to deal with the world of the digital cloud and the Internet of things.

It is hard to exaggerate how difficult it is to conceive and then implement the kinds of restructurings that GE and IBM exemplify. Only some companies attempt the process, and fewer succeed. The losers are dismantled, sold, or disappear. But in a world with fast-changing technology and new global competitors, a firm’s strategy must be reframed and the firm restructured, or it dies.

But still another challenge has emerged to thinking about and managing the resource allocation process. The 1980s saw the emergence of new forms of finance that empowered new corporate strategies. As well, technology proved much more mobile than researchers had previously understood.

The basic microeconomic model of capital budgeting is based on the idea that a firm has a list of project proposals for funding and limited availability of internal capital. The cost of internal capital is calculated as a weighted average of the cost to the firm of equity and debt (the WACC). Capital budgeting systems are still constructed to fund those projects whose projected returns exceed the WACC. Portfolio strategic planning improves the process by making sure that resources are committed only where strategic prospects are approved (which improves the quality of the assumptions in the “business case” underlying proposed uses of funds). But limited capital is a fundamental assumption.

What if capital were readily available? The 1980s saw the development of high-yield debt as a tool to fund entrepreneurial businesses, such as MCI and Turner Broadcasting, but also to fund leveraged buyouts of companies or the divisions of companies. If companies were run with significant overhead and balance sheets with low debt relative to equity, an investor could use debt with a lower cost than the company’s WACC to buy the company for a premium over the stock market price, reduce the overhead, and still make a profit on the investment. As this practice became widespread, firms found their portfolios faced new options in the context of new pressures.

One important option was the spinoff. If a division had good prospects on its own, its management might well want to pursue an independent course without the burden of corporate overhead. From the corporate parent’s perspective, if the division offered no strategic benefits to the other businesses of the company, why not realize the market value of the division, especially if the price offered by the investors who would fund the “management buyout” implied a higher price earnings multiple than the parent enjoyed?

In effect, this new possibility, provided by the very large and liquid U.S. capital markets, changed the strategic challenge to large, multibusiness corporations. A firm’s portfolio of businesses ought to reflect its ability to enhance its prospects because of what some writers call “a parenting advantage” (Goold & Campbell, 1991). A preeminent example of this kind of thinking in practice is the Danaher Corporation. Its remarkable record of profitable organic growth and successful acquisitions has been widely studied. It has devised a strong kit of management tools for operations, planning, and management development that it assiduously applies to its acquisitions and use for organic growth. The “Danaher system” is the corporation’s parenting advantage.

In effect, Danaher thinks through and understands its unique corporate competence. It then manages resource allocation so as to grow in businesses focused on the application of that competence. Many readers will recognize this prescription as one made by Kenneth Andrews (1971) in his foundational book, The Concept of Corporate Strategy. The stark difference is that Andrews could not imagine the life of corporate leaders in the context of contemporary capital markets.

During the same post-1973 oil crisis decades, and reflecting the same liquid financial markets, another force developed that added to the challenge of strategic resource allocation. Starting with Jensen and Meckling’s (1976) development of agency theory, the idea that holders of a company’s common shares actually own the company and that the board and managers of a company are their agents has gained currency. This is not the place to discuss at length what is wrong with this idea—the most important problem is that it does not correspond with corporate law in Delaware and most states—but for boards and top managers, the pressure from the capital market to produce current earnings has become intense. The reason is that activist shareholders have leveraged small holdings—5% or 10% of a corporation’s shares—to impose changes in strategy that improve current share price. They justify their action on the grounds that the objective of all firms is to maximize total shareholder return and that they know how to do this better than self-serving managers.

We have seen important profitable and innovative corporations, such as DuPont, succumb to activist pressures. Returning to the idea of a matrix, the activist seeks to have the low-growth/low-share businesses divested, the high-growth/high-share businesses slow their research and investment, and the risky high-prospect investments largely canceled. The results are higher current earnings and higher share price.

For contemporary managements, the challenge of resource allocation is to fund their strategy in the context of a stock market that seeks high current profitability and growth. Since global growth is slow and overcapacity exists in many parts of the world—especially China—it is hard in this context to think through a program of long-term resource allocation with prospects for sustainable, profitable competiveness.

The reach of financial economics thinking into strategic management has been considerable, with ideas of agency theory and the importance of share price maximization infecting much of the literature. With it has come the idea of delegated decision rights influenced by an appropriate economy of incentives. For a period, it was even fashionable to teach courses in organization using agency theory. All the remarkable progress in understanding the importance of teams and the usefulness of network organization was temporarily discarded as students were taught the importance of aligning decision rights with incentives that paid for performance defined in terms of contribution to total shareholder return.

By way of contrast, work in the emerging field of entrepreneurial management has contributed to our thinking. Ideas about lean start-ups and approaches to funding entrepreneurial strategy have been useful in the context of established corporations thinking through how to think about radical innovation.

Clayton Christensen’s concept of disruptive innovation has been useful (see, for example, Christensen, 1997; Christensen & Bower, 1996). If we focus only on the core idea, the concept helps us to understand how the resource allocation process in an established organization interferes with a company’s ability to fund investment in those new products or business concepts that have low margins in the initial years and product/service features that are inferior using the measures of the incumbent business. The concept has helped companies think through how to carve out such businesses from the ongoing core and allocate resources to them in a separate fashion. Since web-based businesses often have those characteristics, the idea has had a great deal of contemporary applicability.

What is missing is further work on the basic problem of resource allocation. A corporation’s strategy is an evolving target as global markets and technology shift. For example, Olli-Pekka Kallasvuo, past CEO of Nokia and then a star of the strategic management community, noted in a rueful address to the Strategic Management Society that during its strategic management process, Apple was not even on Nokia’s list of competitors 1 year before the iPhone emerged and began to undermine Nokia’s market position. It is hard, but strategic resource allocation requires the willingness to try to “see around corners.” Or as Andy Grove (1996) put it in his book’s title, “only the paranoid survive.”

The reality of resource allocation is that it is a complex process involving technical, economic and financial, organizational, cultural, and interpersonal forces that are fundamentally interrelated. Knowledge about markets, competitors, and technology is dispersed across any corporation. Strategic resource allocation requires conversations that cross functional and divisional lines and levels. Perspectives at different points in the compass are different according to geographic or divisional position. And each of those different perspectives may well be correct.

I believe that it is this complexity that has deterred work on resource allocation. Ideas from many disciplines are needed to describe the process. That means that good research is not always easy to fit in the refereed journals of academia. The sort of descriptive work that has enabled progress certainly does not fit in finance journals. It plays out over long periods of time—generally longer than the time horizon of an academic seeking to publish on a frequent basis. And the data usually defy simple quantification so that abstract modeling and quantitative testing are seldom appropriate research tools. Even more problematic for an academic, the challenge of the task facing corporate management changes significantly as the environment changes. For example, as argued above, the contribution of the growth-share matrix was considerably lessened once technology and capital moved freely in ways that potentially erased the advantages of cumulative experience.

Resource allocation remains the fundamental economic task of a corporate management. All the critical responsibilities of the top management—crafting strategy, building the organization, controlling operations, building and motivating the executive cohort, and managing succession—are involved. What we need badly are new concepts that illuminate how these basic activities influence resource allocation and how those relationships can be improved. There is lots of work to do.