ASQ Research Discussion - Volume 1, Number 2
ASQ Research Discussion Listserver
Volume 1, Number 2, August 25, 1995
Current subscribers: 237
General listserver guidelines:
- Postings and subscription requests: ASQ@UMICH.EDU.
- Please reply to the moderator, rather than all subscribers.
In this issue
- Information about replies to postings
- Responses regarding Will Mitchell's question (in v01n01): What determines whether the likelihood that businesses will fail decreases or increases with age when size is controlled?
1. Information about replies to postings
All new ventures have teething problems. Our first (minor, I hope) problem concerns replies to postings. As you may have discovered, the University of Michigan Business School mail system allows recipients of messages to respond to all members of the mailing list. Unfortunately, this defeats much of our attempt to combine responses in a digest and thereby limit the number of messages that subscribers to the ASQ listserver receive. We are trying to find a way of revising the software. Until we do, we would greatly appreciate if you would reply only to the ASQ@UMICH.EDU account, rather than to all members of the list. Thank you for your help in this effort.
2. Responses regarding Will Mitchell's question (in v01n01): What determines whether the likelihood that businesses will fail decreases or increases with age when size is controlled?
Responses from
- A. Rikki Abzug, rabzug@stern.nyu.edu
- B. Alan D. Bauerschmidt <ALANB@darla.badm.sc.edu>
- C. Te-Chao Chen <ithlib@gate.sinica.edu.tw>
- D. tdelaug@garnet.acns.fsu.edu (Tom DeLaughter)
- E. Henry W. Chesbrough <chesbrou@haas.berkeley.edu>
- F. Thomas Douglas <PA58287@UTKVM1.UTK.EDU>
- G. J Rajendran Pandian <jrp@st-andrews.ac.uk>
A. Rikki Abzug, New York University, rabzug@stern.nyu.edu
Here's hoping that this missive gets appropriately monitored and directed. My first (and at this point, only) reaction to Will Mitchell's query is that Credit Unions and "businesses" might be very different species which react to their environments in different ways. Specifically, Credit Unions are usually 501 (c) (14) tax exempt entities which would place them squarely in the nonprofit world. While nonprofit organizations do indeed evolve (change, die) over life courses (see M. Wood's piece in Nonprofit Management & Leadership), they may face a very different (dependent, even) environment than other "businesses." All of this is not to suggest a necessary sectoral difference, but rather offer the possibility that the difference in findings between the two papers may be a result of the specific "industries" the authors chose to study. That one industry falls into the nonprofit sector may be just a coincidence--or may turn out to have much greater salience if the funding environment, or governance structures of the organizations turn out to be significantly related to the death variables. After putting in my two cents, I'll leave it up to other institutionalists to decide whether organizational field may be the culprit here.
B. Alan D. Bauerschmidt, University of South Carolina <ALANB@darla.badm.sc.edu>
One can wonder if holding size constant is the best way to tease out the influences of age on the survival of organizations. As allometric studies in the biological science advise, a powerful analysis results from correlations among age, size, and life histories. Haldane and S.J. Gould provide early appreciation of the profound connections and Calder's (1984) *Size, function, and life history* may be the definitive work, complete with the allometric formulary. Perhaps, we should not be too quick to put the fine shading of the size variable aside when these further understanding.
C. Te-Chao Chen <ithlib@gate.sinica.edu.tw>
I don't read Mitchell and BWH's papers yet. However their results are very interesting. That inspire me to share one idea that might be relate to their research. Prahalad & Bettis(1986) contended the idea of "dominant logic" within performance and diversification. They emphasized the role of top management. Such a variable(if we can give it an operational definition) may be consider into such a topic.
Besides, many Japanese business groups(e.g. Sumitomo, etc) may be use as another sample in further studying that may be give us a more global idea in this research.
That just my rough idea.
D. Tom DeLaughter, Florida State University College of Business, tdelaug@garnet.acns.fsu.edu
May I confound the question by distinguishing between organizational age and product age. The life cycle of products is well researched. In the samples you mentioned, the medical instrument/technology field changes products frequently as changes in the environment (market, technology, etc.) dictate whereas credit unions have not changed products as rapidly as changes in the marketplace because the environment (regulatory) dictates otherwise. If credit unions show a positive correlation between institution failure and age, while the medical industry sample showed the reverse, it may be because of product age (decline phase of product). I think that it would be worth while to add product age to the model to see what happens.
E. Henry W. Chesbrough, UC Berkeley <chesbrou@haas.berkeley.edu>
I wanted to submit one response to your question about whether age increases or decreases the likelihood of business failure, once age is controlled.
My response stems from preliminary work I am doing on a particular industry in the US and in Japan. My initial observations are that there may be different answers to your question in each country, in part for the reasons you mentioned (environment supports divestiture, and organizational routines remain of value).
More specifically, I observe that the labor market and capital market in Japan respond to "legitimacy" questions far longer than may be the case in the US. Top engineering graduates in Japan, for example, often are courted by many leading companies, and these companies in turn maintain relations with many of the students' professors. These ties take significant time to form, and are likely to be quite slow to change.
Similarly, the main bank financial system supplies a variety of formal and informal control mechanisms for evaluating and overseeing capital lent to companies, and many of these informal controls (posting of managerial personnel, for instance, or reciprocal shareholdings) rise with age and are likely to be slow to change. And of course, "failure" in Japan may not lead to dissolution, at least not for larger companies. Rather, the distressed firm's creditors may step in and replace the management (thereby salvaging useful routines, perhaps) while preserving the operation of the firm.
As we move to more careful research about the effects of aging on business dissolution, I hope we are able to incorporate evidence from non-US settings. Where different empirical patterns emerge, we then need to inquire about the source(s) of the different patterns, and our theorizing will be stronger as a result.
F. Thomas Douglas, University of Tennessee, <PA58287@UTKVM1.UTK.EDU>
With respect to your question concerning the relationship between business failure and business age, controlling for size, I wonder if such an analysis suffers from a lack of dynamics and this is the cause of the seemingly inconsistent results. Barnard (1968: 251-252) discusses efficiency and wonders if firms can be efficient if they are not growing. Chandler (1962) also discusses growth in terms of expanding into new markets and product lines in order to ensure the full use of resources, which also seems to be an efficiency argument. If we hold size constant in our analysis of business failure and attempt to estimate its relationship with organization age, are we leaving out a possible interaction between age and size that has been argued as potentially necessary for efficiency and for the long term survival of the firm?
G. J Rajendran Pandian <jrp@st-andrews.ac.uk>
Size affects the probability of failure due to increased inertia and possibly the refusal to adapt. But, this relationship need not be linear and also as Rikki pointed out could differ across industries. The nonlinearity would mean that controlling for size as WM did might (!) not be enough. Also, the average size of the sample could be very different in these two studies and hence we probably cannot compare.
Age also affects the probability of failure to a great extent the same way (develop a strong culture and managers refuse to see reality). The liability of newness argument does not suggest that the failure should reduce with age (except when the environment is ever supportive).
Also, the nature of the market is very important to suggest whether or not we can compare the findings as we attempt to (Just like Chung-chi Hung, I have not read both these articles yet!) and I do not know anything about both the markets for me to make intelligent guesses.
Hope to give more comments after I read these articles!
