Search

The number of private equity (PE) firms that have started to use artificial intelligence (AI) in investment decisions has risen rapidly over the past 10 years. This article provides a detailed account that can serve as a template for others in the industry who wish to make better investment decisions using AI. The news is both good and bad. The increased use of AI in PE and venture capital will greatly increase operational efficiency and transform the ways in which partners perform their work. It will allow for the entry of new firms but will also lead to a technological arms race and is predicted to cause an eventual industry shakeout. TOPICS:Private equity, big data/machine learning, performance measurement Key Findings ▪ The number of private equity firms that have started to use AI is rising rapidly. ▪ Use of AI will greatly transform the deal-making process. ▪ By increasing efficiency, AI will likely cause an industry shakeout.

We document that since 1997, the rate of startup formation has precipitously declined for firms operated by U.S. PhD recipients in science and engineering. These are supposedly the source of some of our best new technological and business opportunities. We link this to an increasing burden of knowledge by documenting a long-term earnings decline by founders, especially less experienced founders, greater work complexity in R&D, and more administrative work. The results suggest that established firms are better positioned to cope with the increasing burden of knowledge, in particular through the design of knowledge hierarchies, explaining why new firm entry has declined for high-tech, high-opportunity startups.

This paper contains the on on-line appendix to the paper with the same title forthcoming in ILRR.; Is the Bayh-Dole intellectual property regime associated with more and better academic entrepreneurship than the Professor Privilege regime? The authors examine data on US PhDs in the natural sciences, engineering, and medical fields who became entrepreneurs in 1993–2006 and compare this to similar data from Sweden. They find that, in both countries, those with an academic background have lower rates of entry into entrepreneurship than do those with a non-academic background. The relative rate of academics starting entrepreneurial firms is slightly lower in the United States than in Sweden. Moreover, the mean economic gains from becoming an entrepreneur are negative, both for PhDs originating in academia and non-academic settings in both countries. Analysis indicates that selection into entrepreneurship occurs from the lower part of the ability distribution among academics. The results suggest that policies aimed at supporting entrepreneurial decisions by younger, tenure-track academics may be more effective than general incentives at increasing academic entrepreneurship in general.

This paper assesses the relative importance of the complementary assets and financial capital that business partners may add to the original inventor-entrepreneur. Projects run by partnerships were five times as likely to reach commercialization as those without business partners, and they had mean revenues approximately 10 times as great as projects run by solo entrepreneurs. These gross differences may be due both to partners impacting business success that is, who the particular partners were, and to selection of the type of project or of whom to select as a partner. After controlling for selection effects and observed/unobserved heterogeneity, the smallest estimate of partners' complementary assets approximately doubles the probability of commercialization and increases expected revenues by 29% at the sample mean. Our findings suggest that a critical policy option to increase commercialization rates and revenues for early-stage businesses is to support the market for finding skilled partners.

Using a dataset with over 24 million year-employment observations and the universe of more than 230,000 entries into entrepreneurship in one economy we analyze the gross (including the founders), net (excluding the founders), and new (jobs to the former unemployed or those outside the labor force) job creation by entrepreneurs two and six years after start-up. These novel measures of job creation show that the average entrepreneur does not create any jobs for any other than him/her-self, and typically arrives from having another job. Thus, short term job creation by entrepreneurs involves a reshuffling of jobs from older to new firms rather than creating new jobs.

Do equity crowdfunding investors herd? We build a model where informed and uninformed investors arrive sequentially and choose whether and how much to invest. We test the model using data of investments on a leading European equity crowdfunding platform. We show theoretically and find empirically that the size and likelihood of a pledge is affected positively by the size of the most recent pledges, and negatively by the time elapsed since the most recent pledge. The empirical analysis is inconsistent with naive herding, independent investments, or exogenously correlated investments. Investors rationally herd but social learning about the project's quality stops only when herding induce investors not to invest.

We debate the motivation for and effectiveness of public policies to encourage individuals to become entrepreneurs. Reviewing established evidence we find that most western world policies do not greatly reduce or solve any market failures but instead waste taxpayers’ money, encourage those already intent on becoming entrepreneurs, and mostly generate one-employee businesses with low growth intentions and a lack of interest in innovating. Most policy initiatives that would have the effect of promoting valuable entrepreneurship would not be recognizable as such, because they would primarily address other market failures: a central-payer healthcare would remove health-care related distortions affecting employment choices; greater STEM education would produce more engineers of which some start valuable new firms; and labor market reform to encourage hiring immigrants in jobs they have been educated for would reduce inefficient allocation of talent to entrepreneurship.

We debate the motivation for and effectiveness of public policies to encourage individuals to become entrepreneurs. Reviewing established evidence we find that most western world policies do not greatly reduce or solve any market failures but instead waste taxpayers’ money, encourage those already intent on becoming entrepreneurs, and mostly generate one-employee businesses with low growth intentions and a lack of interest in innovating. Most policy initiatives that would have the effect of promoting valuable entrepreneurship would not be recognizable as such, because they would primarily address other market failures: a central-payer healthcare would remove health-care related distortions affecting employment choices; greater STEM education would produce more engineers of which some start valuable new firms; and labor market reform to encourage hiring immigrants in jobs they have been educated for would reduce inefficient allocation of talent to entrepreneurship.

International audience; We examine the accuracy of forecasts of the commercial potential of new product ideas by experts at an Inventor's Assistance Program (IAP). Each idea is evaluated in terms of 37 attributes or cues, which are subjectively rated and intuitively combined by an IAP expert to arrive at a forecast of the idea's commercialization prospects. Data regarding actual commercialization outcomes for 559 new product ideas were collected to examine the accuracy of the IAP forecasts. The intensive evaluation of each idea conducted by the IAP produces forecasts that accurately rank order the ideas in terms of their probability of commercialization. The focus of the evaluation process on case-specific evidence that distinguishes one idea from another, however, and the corresponding neglect of aggregate considerations such as the base rate (BR) and predictability of commercialization for new product ideas in general, yields forecasts that are systematically miscalibrated in terms of their correspondence to the actual probability of commercialization. Copyright © 2007 John Wiley & Sons, Ltd.

We model the decision making process used by Experts at the Canadian Innovation Centre to classify early stage venture proposals based on potential commercial success. The decision is based on thirty-seven attributes that take values in { - 1 , 0 , 1 } . We adopt a conjunctive decision framework due to Astebro and Elhedhli (2005) that selects a subset of attributes and determines two threshold values: one for the maximum allowed negatives (n) and one for minimum required positives (p). A proposal is classified as a success if the number of positives is greater than or equal to p and the number of negatives is less than or equal to n over the selected attributes. Based on a data set of 561 observations, the selection of attributes and the determination of the threshold values is modeled as a large-scale mixed integer program. Two solution approaches are explored: Benders decomposition and Tabu search. The first, was very slow to converge, while the second provided high quality solutions quickly. Tabu search provides excellent classification accuracy for predicting commercial successes as well as replicating Experts’ forecasts, opening the venue for the use of Tabu search in scoring and classification problems.

This is the on-line appendix to the paper "Business Partners: Complementary Assets, Financing, And Invention Commercialization" published in Journal of Economics and Management Strategy, The paper assesses the relative importance of the complementary assets and financial capital that business partners may add to the original inventor-entrepreneur. This appendix contains the theoretical model, details on the sampling process, demographics, alternative estimation results and other details.

We use creativity theory to analyze the effects of occupational job variety and industry variety on invention quality, and entrepreneurial earnings. We test our ideas with survey data from 770 inventor–entrepreneurs who commercialized their own inventions. Results suggest that occupational and industry variety substitute for each other in positively affecting invention quality whereas a lack of industry variety is associated with greater entrepreneurial earnings. Results are consistent with the idea that high levels of both occupational and industry variety enables the generation and discovery of inventions, but these ideas are usually not technically feasible or financially viable.

We explore whether the Bayh-Dole intellectual property regime is associated with more, and more valuable academic entrepreneurship than the “Professor’s Privilege” regime. Using data on U.S. STEM Ph.D.’s becoming entrepreneurs during 1993-2006 and similar data from Sweden we present evidence showing that in both countries the entry rate into entrepreneurship is lower for those originating from academia than for those orginating from non-university employment, and that the relative rate of academic entrepreneurship is slightly lower in the U.S. than in Sweden. We also find that the mean economic gains for becoming an entrepreneur is negative, both for Ph.D.’s originating from academia and non-university alike in both countries. Further analysis indicates that in both countries there is selection from the bottom of the ability distribution among academics. The results suggest that policies aimed at screening entrepreneurial decisions by younger, tenure-track academics may be more effective than general incentives to increase academic entrepreneurship.

1,095 responses were received from a telephone survey of independent inventors. The majority of inventors surveyed (89%) are male, and a plurality of their inventions are designed for consumer products (47%). Conditional on commercializing their invention, and at development costs about 1/8 of those in established firms, inventive efforts by independent inventors lead to gains virtually comparable to those of established firms. Their innovations survive for about as long as the average start-up. Gross profit margins are comparable to the pharmaceutical industry (29%). However, only a small fraction of inventions developed by independent inventors reach the market (6.5%). The probability of reaching the market is four to eight times less than for inventions developed by established firms. Why do only a fraction of inventions developed by independent inventors become commercialized when those that do commercialize are quite profitable and survive for as long as other start-ups? (Reprinted by permission of the publisher.)

Recent evidence comparing earnings from entrepreneurship versus wage earning shows that, after allowing for obvious observable differences, most entrepreneurs in most developed countries earn less than similar wage-earning employees. Does this mean that the decision to become an entrepreneur should be discouraged? The answer depends in part on whether we believe that entrepreneurs report their income truthfully or not. Adjusting for what is considered to be underreporting by entrepreneurs lifts entrepreneurial earnings by between 10% and 40%, reversing the fortunes of the entrepreneur such that they appear to be earning much more than their counterparts in a wage-earning job. If this adjustment should prove to be appropriate, then there is no obvious reason to increase the incentive for individuals to become entrepreneurs (such as with tax breaks or direct start-up subsidies) in developed countries, and there is reason, instead, to discuss decreasing these subsidies.

International audience; Can management affect the rate by which organizational members adopt a new computer based communication medium? The question is addressed by comparing the rate by which employees start to use an electronic mail system (EMS) and case descriptions across four departments during a five year period in a large Swedish manufacturing company. Data on EMS use are fitted to a model describing the spread of use as a function of social and management influence. Data are consistent with the hypothesis that the rate of new users of EMS can be influenced by both department-level management and peers. Faster growth in use occurred when department-level management exercised implementation efforts. The rate of growth increased significantly when a local information center (IC) was established. The more resources the local IC was supplied with the higher the rate of growth in EMS use. Management of the implementation process may be especially preferred when introducing end-use network technologies in organizations due to interdependent demand among users

We study the impact of a new nationally advertised six-month intensive training program to encourage social entrepreneurship among youth. Program costs were on the order of 12,000 euros per partici...

We develop a new test of simultaneous versus sequential complementarities by using a discrete-time binary model for survival-type panel data on the adoption of one, both, or neither of two technologies while controlling for unobserved heterogeneity by instrumenting key variables. We examplify using data on the adoption of CAD and CNC technologies by US metalworking plants. Results indicate significant complementarities. Simultaneous adoption is in this case more likely than adoption of the two technologies in isolation. Sequential adoption is significant but in only one direction. Prior adoption of CNC has a positive effect on the posterior adoption of CAD. We find no evidence of the opposite.

The topic of Entrepreneurial Finance involves many issues, including but not limited to the risks and returns to being an entrepreneur, financial contracting, business planning, capital gaps and the availability of capital, market booms and busts, public policy and international differences in entrepreneurial finance stemming from differences in laws, institutions and culture. As these issues are so extremely broad and complex, the academic and practitioner literature on topic usually focuses on at most one or two of these issues at one time. The Oxford Handbook of Entrepreneurial Finance provides a comprehensive picture of issues dealing with different sources of entrepreneurial finance and different issues with financing entrepreneurs. The Handbook comprises contributions from 48 authors based in 12 different countries. It is organized into seven parts, the first of which introduces the issues, explains the organization of the Handbook, and briefly summarizes the contributions made by the authors in each of the chapters. Part II covers the topics pertaining to financing new industries and the returns and risk to being an entrepreneur. Part III deals with entrepreneurial capital structure. Part IV discusses business planning, funding and funding gaps in entrepreneurial finance with a focus on credit markets. Part V provides analyses of the main alternative sources of entrepreneurial finance. Part VI considers issues in public policy towards entrepreneurial finance. Part VII considers international differences in entrepreneurial finance, including analyses of entrepreneurial finance in weak institutional environments as well as microfinance.

We study the effects of three cognitive biases by the entrepreneur on obtaining funding. We find planning fallacy to increase funding amounts, whereas optimism and overconfidence by the entrepreneur have no effects on funding amounts from others. Further, planning fallacy positively impacts the probability of strong-tie (inside) investments but negatively impacts the probability of weak-tie (outside) investments. Mediation analyses further show that planning fallacy positively impacts venture performance through both self and other-investor funding amounts. Our findings are not consistent with the pecking order theory of informal finance, and suggest positive effects of at least one cognitive bias on entrepreneurial business success through increased funding.

International audience; Lazear (2005) suggests that entrepreneurs should be generalists, while those who work for others should be specialists. Many prospective entrepreneurs will develop varied skills by engaging in a variety of employment activities prior to becoming an entrepreneur, and incomes are higher for those that do so. An alternative view predicts that those with greater taste for variety are more likely to become entrepreneurs. Varied employment prior to becoming an entrepreneur is simply an expression of this taste, and is associated with lower earnings. Data from a survey of 830 independent inventors and 300 individuals from the general population are used to discriminate between these two theories. The results show that inventor-entrepreneurs typically have a more varied labor market experience, and that varied work experience is associated with lower household income.

This paper studies the effect of business partners on the commercialization of invention-based ventures, and it assesses the relative importance of the complementary skills, contacts, and financial capital that these partners may add to the original inventor-entrepreneur. Projects run by partnerships were five times as likely to reach commercialization as those without partnerships, and they had mean revenues approximately ten times as great as projects run by solo entrepreneurs. These gross differences may be due both to partners impacting business success that is, who the particular partners were, and to selection of the type of project or of whom to select as a partner. After controlling for selection effects and observed/unobserved heterogeneity, the smallest estimate of partners’ skills and contacts approximately doubles the probability of commercialization and increases expected revenues by 29% at the sample mean. Our findings suggest that a critical policy option to increase commercialization rates and revenues for early-stage businesses is to support the market for finding skilled partners.

Earlier research on the role of universities in fostering entrepreneurial economic development almost exclusively covers spin-offs by faculty and staff. In contrast, we provide general evidence from the U.S. showing that the gross flow of science and engineering university graduates’ start-ups is at least an order of magnitude larger than faculty spin-offs, that a recent graduate is twice as likely as her Professor to start a business within three years of graduation, and that the graduates’ spin-offs are not of low quality. Three case studies illustrate how universities may stimulate science and engineering students and recent graduates to create new firms of high quality. We conclude that transforming university goals and practices toward increasing start-ups led by faculty might not be the most effective way for universities to stimulate entrepreneurial economic development.

Direct individual-specific evidence on the profit motives of entrepreneurs has been lacking. Indeed, recent work suggests that entrepreneurship may be a non-profit-seeking activity, and that entrepreneurs evaluate risk oddly. In contrast, using direct prior measures of economic potential for inventions, we find that inventors respond sensibly to pecuniary incentives. We construct an empirically-relevant economic theory, with heterogeneous inventions and inventors, outside opportunities, sunk and non-sunk costs, and risk, to guide our investigation of the data. We collected assessment data from a center paid by inventors to assess their inventions' economic potential, and followed up with surveys to determine concrete economic outcomes. We find that inventors decide whether to commercialize their inventions (and later whether to exit) in a manner consistent with profit-seeking motives and risk aversion.

This article examines the impact of key success factors on the survival of innovations that have reached the market and were developed by inventors outside of established organizations. It is of interest to learn which characteristics predict, at an early stage, the duration of the innovation’s length of sales, because this duration is important to the financial success of new products. A focus on survival also can contribute conceptual clarity to the study of new product development. This study uses the Inventor’s Assistance Program (IAP) at the Canadian Innovation Centre (CIC) in Waterloo, Canada, as the source of data. The CIC is a not-for-profit agency that provides various services to foster business development involving innovative products and services. Analysts in the IAP evaluate a specific product idea or invention on 37 dimensions before it has reached the market. The data for the present study involved these 37 variables evaluated each with a three-point linguistic scale. As the evaluations of the criteria are subjective, they might be argued to contain inaccuracies compared to objective data. On the other hand, the analysts use multiple related measures of concepts that have been shown to increase predictive accuracy. The use of experts who are unrelated to the projects avoids decisionmaking biases potentially associated with project managers’ assessment of their own projects, such as unrealistic optimism. The recording of the expert evaluations of the ideas before they reached the market and independent of the measure of success, rather than using post-project completion evaluations, eliminates measurement biases such as hindsight bias and common method variance bias. Identifying information was used in these records to conduct a telephone survey of the inventors. An exploratory method of data analysis is identified and used that distinguishes research-appropriate constructs and their indicators in these data. Cluster analysis was performed, and survival regression correlated cluster scores with survival. Three variables were found to significantly affect survival: anticipated stable demand, price required for profitability, and technical product maturity. In addition, the degree of competition had a marginally significant effect. Because these variables can be assessed at an early stage of an inventions’ development, the expected survival time for a specific invention may be computed by entering these assessment values into the described survival model. Then this and other information may be used to compute the expected return of an invention.

We analyze the simultaneous diffusion of multiple process technologies that are related. A new econometric model is used to examine the presence of complementarities, testing for strong one-step-ahead non-causality and strong simultaneous independence. Results indicate significant complementarities between CAD and CNC technologies. Prior adoption of either of the two technologies has a large effect on the posterior adoption of the other one; in addition, simultaneous adoption is found to be more likely than adoption of the two technologies in isolation. Consistent with the presence of complementarities, we also find evidence of substantial price cross-effects: a decrease in the price of CAD (or CNC) increases the adoption probability of CNC (or CAD). Lastly, the increase in the likelihood of adopting the complementary technology turns out to depend on several plant-specific moderating factors.

International audience; I propose that sunk costs of learning and the output over which these costs are spread determine the probability and depth of technology adoption. Depth of adoption describes the extent to which firms exploit the advantages of the technology. I find that plant size but not firm size predicts CAD and CNC adoption. Learning costs are lumpy, are closely connected to technology adoption and determine both the probability and depth of adoption. Depth of adoption is considerably more plant idiosyncratic than the decision to adopt.

The impact of 36 innovation, technology, and market characteristics on the probability that early stage R&D projects will reach the market is examined. Analysis is based on data from 561 R&D projects developed by technological entrepreneurs. Four characteristics stand out as most predictive: expected profitability, technological opportunity, development risk, and appropriability conditions. These predict future commercial success well with an out-of-sample accuracy of 80.9 %. This model performs better than R&D managers' and venture capital's (VC's) forecasts of success and has the potential to be used as a screening tool for early stage R&D investment reviews by, for example, VCs. Implications for both research and practice are discussed.

International audience; We investigated the relation between the survival of new small businesses and bank loans. This was done using a model that included other loan sources, human capital variables, and company and industry descriptors. We found there is a negative correlation between having a bank loan and business survival, and a positive correlation between having a non-bank loan and survival. However, having a bank loan was a ceteris paribus positive predictor of the survival of start-up companies. Our findings enabled some inferences about the process of loan source selection by start-up business owners, and about the banks' loan granting process.

We investigate the decision heuristics used by experts to forecast that early-stage ventures are subsequently commercialized. Experts evaluate 37 project characteristics and subjectively combine data on all cues by examining both critical flaws as well as positive factors to arrive at a forecast. A conjunctive model is used to describe their process which sums good and bad cue counts separately. This model achieves a 94.1% classification accuracy of the experts' correct forecasts of 561 projects. The model correctly predicts 86.0% of outcomes in out-of-sample, out-of-time tests. Results indicate that reasonably simple decision heuristics can perform very well in a natural and very difficult decision-making context.