Estimating the Small Business Failure Rate: A Reappraisal

Abstract

Small business research in Australia has become increasingly concerned with the phenomenon of business failure. This is also the case for research into other types of business; but in the case of small business there is a preoccupation with the apparently high rates of collapse. Although it is the causes of failure which are ultimately of most importance, the extent of failure itself is a measure which has received priority status in much of the literature. Many researchers suggest that continued research, and greater academic resources be utilized in attempting to produce statistics to indicate the magnitude and timing of failure. Yet, in Australia, there continues to be few attempts to make such assessments. One reason for this may be that accurate or meaningful assessments are extremely difficult to make. Problems of business classification, definition of failure, data collection, and statistical analysis provide underlying difficulties. This note reviews the extant studies in Australia, and concludes that the line of research has probably proceeded as far as it can, in producing a general case for small business failure rates in Australia. Although some indication of the of failure may be appropriate, attempts to increase the precision of these statistics may be neither fruitful nor helpful. In particular, failure statistics which enable broad cross-industry comparisons will be more useful than attempts to specify accurately the actual of failure. Instead, it is suggested that resources be directed to further research into the causes of failure. Failure Rates Discussions of small business failure commonly include amorphous sets of numbers, intended to emphasize the enormity of the problem. Frequently, statements can be discovered suggesting that x of small businesses fail within y years. These percentages are usually high. . . . half of all small businesses fail within the first two years and 80 within the first five years (The National Times, August 1980). . . . about of companies which start out with nothing will fail within two years.' The intention may be laudable: to emphasize that the owner/managers of undercapitalized businesses with inadequate expertise, run into problems. But in fact, rates of failure are often skewed because of variances among industry sectors, and generic figures, such as 70 percent failure may be more misleading than useful. The inherent problem is: How accurate a measure of these variances, or other statistical properties, can really be obtained? This problem can be broken down into several questions. First, is it possible to arrive at a meaningful rate of failure, even supposing that the were partitioned for industry sectors? Is one study which purports to supply double-digit percentages (e.g., 69 percent failure), going to be more relevant than that of another study which may give a quite different figure? How useful is the figure itself? To answer such questions, it is necessary to examine the literature which has already attempted such estimations. Price conducted a study of closures among manufacturing firms in the State of Victoria between 1966 and 1977. The average closure for small manufacturers (fewer than one hundred employees) was approximately 3.0 per year. Price estimates the net closure for non-manufacturing firms to be approximately 3.5 percent. This result conflicts with several other failure studies. For example, Williams conducted a longitudinal study spanning the years 1973 through 1985. The average of failure within the first year of operations was 35.4 percent, while for the first three years the average failure was 47.4 percent. In contrast, the Institute of Industrial Economics (Newcastle University-NIE), conducted a study in 1978 of 2,563 small businesses, which had been the subject of an earlier study in 1973. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 123 citations exist upstream.