Most entrepreneurs -- along with those who study them -- assume that businesses become more valuable as they grow. This assumption draws support from studies of experience curves (e.g., Abernathy and Wayne 1974), which indicate that unit manufacturing costs decrease with accumulated volume. These cost savings may come from learning how to use resources more productively. The PIMS literature (e.g., Buzzell and Gale 1987) extends this idea, arguing that higher volumes imply higher market shares and, hence, higher profits. The "liability of newness" concept includes small size as one reason for the high risk of failure of new businesses; as these businesses mature and grow, they begin to enjoy economies of scale, such as quantity discounts and more efficient use of people and equipment.

In contrast to these linkages between increased volume and increased wealth, anecdotal evidence reveals that increases in size are not always related to increases in net worth. A successful Minneapolis bookseller who opened a second store became insolvent from higher costs and unexpectedly low additional demand, and had to close both stores. In the same city, a restaurant chain expanded from one location to four, always in hope that greater volume would improve profits and cashflow; the stores closed one by one until the chain became bankrupt. If growth routinely led to greater wealth, Rappaport (1986) would not need to remind us that greater size does not always lead to greater shareholder value.

Microeconomic theory offers a way to reconcile this apparent inconsistency. Theory states that a firm will maximize profits if its output stands at the level at which marginal costs equal marginal revenue (e.g., Thompson 1989). Studies of individual firms (see, for example, "The University of Minnesota Press," in (Cardozo, 1979)) show the existence of "local optima," output levels that maximize profitability within particular volume ranges. These "local optima" occur because planned cost increases take place in a stepwise fashion. A "local optimum" typically appears just before a step up in costs, for example, just before adding employees, equipment or facilities. Exhibit 1 shows what a set of "local optima" might look like for an emerging firm.

In Exhibit 1, it would pay for a firm to grow up to output volume 01, after break-even, profits would increase steadily up to that volume. It would not, however, pay for a firm to grow beyond 01 unless the firm could grow to volume 02 (where profits are equal to those at 01) or higher. At any level of volume between 01 and 02, profits are lower than they would be at 01 or 02. Because profits are equal at 01 and 02, the firm will be no better off at 02 than it was at the smaller volume of 01. In fact, since more resources are likely to be needed to produce at the higher volume of 02 than at 01, the firm’s return on resources will likely be lower at 02 than at 01. If the firm can grow beyond 02 to 03, it will become more profitable; but growth beyond 03 will again reduce profits.

Literature on new independent businesses indicates that those businesses start small and generally remain in the same relative size ranking compared to others in their cohorts as the businesses mature (Cardozo, et al, 1991). We expect that most new businesses will begin operation near the left of the volume scale in Exhibit 1, and become steadily more profitable as they increase in volume. Should they "overshoot" the optimum indicated by 01, they may seek to reduce their operations to increase profits, or else seek substantial additional resources to increase volume beyond 02. The latter course may be difficult, as low or declining profits make investors and lenders reluctant to commit the resources needed for significant growth; in that instance, growth plans are likely to be curtailed. Finally, some businesses may seek to grow only to the point at which they earn a satisfactory profit, and will not expand beyond that point (Cardozo, et al, 1992). In general, then, we would expect (H1) businesses to show increases in net worth with increases in sales.

While volume of the smallest businesses may be approaching 01, the largest businesses may be growing from 02 to 03 (or beyond). Businesses in between them in size might be growing from 01 to 02. If this were the case, we should expect (H2) the largest businesses to exhibit the largest increases in value per increase in sales; the smallest businesses to show the second-largest increase in value per increase in sales; and mid-size businesses to show the lowest increases in value per increase in sales. The values of "b" in the correlation or regression equation "y = a + bx" should be highest for the largest businesses, next for the smallest businesses and lowest for medium-size businesses.

We should also expect (H3) that the effects of changes in volume on changes in profitability would be greatest in industries in which higher proportions of costs were fixed, i.e., in industries in which the value of adding a customer was greatest (at the extreme, a utility fits this definition).

If increases in sales are correlated with increases in net worth, as predicted by H1, we should expect to observe increases (H4) in operating efficiencies (profit divided by assets, or return on assets (ROA)).

In sum, we hypothesize the following:

H1: Increases in sales will be positively related to increases in net worth.

H2: Values of beta in correlation/regression equations will be highest for the largest businesses, next highest for the smallest businesses, and lowest for businesses intermediate in size.

H3: Changes in net worth for every dollar change in sales will be higher for manufacturing firms than for service firms.

H4: Increases in volume will be positively correlated with increases in operating efficiency (profit/assets).



To explore relationships between growth in size and increases in wealth among new businesses, survey data were collected in 1992-93 from 576 startup firms in Minnesota and Pennsylvania. These firms, all of which were born between 1979 and 1984, were part of a sample initially drawn in 1986-87 (see Reynolds and Miller (1988) and Reynolds, et. al (1984) for discussions of how the sample was built) that responded to a follow up questionnaire in 1992-93. Since the number of firms responding to the follow up in 1993 was substantially smaller than the number who completed the initial questionnaire in 1986-87, this follow up group represents some of the most resilient firms from the original sample. For example, less than 9 percent of the 576 follow up respondents were in the retail or restaurant industry, which are usually the industries which account for the highest percent of new business startups and also have the highest rate of failure. Almost a quarter of these firms were manufacturing firms, another quarter were in distributor services, and just under a third were involved in producer services. In common with most surveys of new firms, more than 80 percent of these firms had fewer than 10 employees; fewer than 15 percent displayed consistent double-digit growth.

The 1992-93 follow up survey contained a matrix of questions asking firms to provide baseline financial data--total sales, after tax profits, year end debts and year end assets--for each of their last seven years of operation. At the time of the survey, these years were from 1985 through 1991. Since these questions cannot be answered through simple recall, and since many firms refuse to make such financial information public (in spite of assurances of anonymity and confidentiality), only 118 of the 576 respondents provided full financial data for each of the seven years. For reasons of convenience, as well as lack of evidence to suggest that those firms who provided full financial data are dramatically different from those who did not, those 118 firms provided the basis of this study.

For this study, we have defined changes in size as changes in delivered output, or sales. We chose not to use physical production, both because it is difficult to compare "units" among firms and because production stored as inventory does not generate cash profits. We defined changes in wealth or value of the business in terms of change in net worth over a specified period. The cumulative change in net worth is a more stable indicator than annual profits, which vary greatly from year to year. We recognize that net worth may be understated to the extent that entrepreneurs withdraw funds (other than salary) from the business for personal use. We measure operating efficiency as profit divided by assets, or return on assets (ROA).

All dollar figures presented are in 1983 dollars to remove inflation effects. These figures can be converted approximately to 1996 dollars by multiplying them by 1.5.

Previous Page | Main Menu | Next Page


1997 Babson College All Rights Reserved
Last Updated 1/15/97 by Geoff Goldman & Dennis Valencia

To sign-up for the Center for Entrepreneurial Studies' publication lists,
please register with the
Entrepreneurship WebTeam.