The NBER Reporter 2008 Number 1: Conferences
Ninth Annual Conference in India
International Differences in Entrepreneurship
Location of Biopharmaceutical Activity
Incentive and Distributional Consequences of Tax Expenditures
The New World of Private Equity
Velez and his co-author characterize entrepreneurship in developing economies through a case study for Colombia. They document self-employment and business ownership since the 1980s; the relative size of these groups within the labor force is stable across time, but they differ significantly in important observable dimensions, such as education and business sector. Next the authors study the motivations for becoming an entrepreneur. They analyze the transition into and out of potential forms of entrepreneurship by measuring the flows across occupations, and study the determinants of entry and exit into and out of self-employment and business ownership. They find surprisingly little transition between self-employment and business ownership. Second, they focus on the financial motivations by measuring the differences in earnings from self-employment and business ownership relative to salaried work, at the mean and along the distribution. There is a substantial earnings premium to becoming a business owner, but it is not financially attractive to become self-employed. The results of this paper suggest that while business ownership is what the literature associates with entrepreneurship, self-employment is basically a subsistence activity.
Lelarge and her co-authors use information on a French loan guarantee program to assess the consequences of credit constraints for new ventures. Loan Guarantee Programs, as implemented in France, are an effective instrument for helping young firms grow faster, in terms of both employment and capital. These effects are quite persistent, because they are still significant four years after obtaining the guarantee. Loan guarantees also allow firms to pay cheaper interest rates, but a potential drawback of this policy is that guaranteed ventures adopt riskier strategies, and thus filing for bankruptcy occurs more often. Finally, the authors find no effect, at the industry level, on creation rates.
Iyer and Schoar explore the importance of culture in determining contractual outcomes across different communities. They set up an audit study and send buyers (from different communities) to contract with wholesalers (who also belong to differ communities) in the pen industry in India. They find that wholesalers from the most entrepreneurial community in India (Marwari) offer lower prices. Furthermore, wholesalers from Andhra offer higher prices than dealers from other communities. Iyer and Schoar also find that when the buyer and seller from the same community contract, the terms of the contract are more favorable. These results suggest that culture has an important effect on contract terms.
The World Bank Group Entrepreneurship Survey measures entrepreneurial activity around the world. The database includes cross-country, time-series data on the number of total and newly registered businesses for 84 countries. Klapper and her co-authors find significant relationships between entrepreneurial activity and indicators of economic and financial development and growth, the quality of the legal and regulatory environment, and governance. Their analysis shows the importance of electronic registration procedures to encourage greater business registration. These results can guide effective policymaking and deliver new capabilities for identifying the impact of reforms.
Using Census microdata, Fairlie and his co-authors provide the first comparative examination of the education levels, business ownership, and business performance of Asian immigrants in the United States, Canada, and the United Kingdom. They specifically compare the effects of education and other determinants of business ownership and performance in the three countries to help explain the heterogeneity across Asian immigrant groups within countries and across countries. Asian immigrants to all three countries have education levels that are higher than the national average, and in the United States the education levels of Asian immigrants are particularly high relative to the entire population. Some of the variation in the education of Asian immigrants across the United States, Canada, and the United Kingdom is likely attributable to immigration policy. For example, the United Kingdom is more likely to accept immigrants in the refugee or asylee category than the other two countries. The business ownership rates of Asian immigrants in the United States and Canada are similar to the national average, and are substantially higher than the national average, and highest among all three countries, in the United Kingdom. In Canada and the United States, Koreans have high rates of business ownership, while Philippinos have low rates of business ownership. On average, the income of Asian immigrant business owners is only slightly above the national average (in the United States), or below the national average (in Canada), and is thus not the broad picture of success that is often portrayed. In both the United States and Canada, the business income of Indians is high relative to the national average. Estimates from regression models for business ownership, log business income, and employment reveal interesting differences across the three countries. In particular, education is found to be a positive, although not strong, determinant of business ownership in the United States and Canada, but not in the United Kingdom. In the United Kingdom, education has no effect on business ownership. When the authors examine business income, they find large, positive effects of education in the United States and Canada. In the United Kingdom, they find smaller positive effects of education on employment. The findings for education imply that the relatively high levels of education among some Asian immigrant groups do not have a large influence on business ownership rates for the groups, but have a large effect on business performance at least in the United States and Canada. In regression models for business ownership, the coefficients on Asian immigrant groups generally do not change after controlling for education and other demographic characteristics. In contrast, there are large changes in coefficients for log business income in the United States and Canada after controlling for education and other variables, suggesting that education differences are important. Decomposition estimates indicate that high levels of education contribute to higher business income levels among Indians and Pakistanis in the United States. Another interesting finding from the analysis is that Asian immigrants even from the same source country are generally much more educated in the United States than in Canada or the United Kingdom. For example, 76.2 percent of Indian immigrants in the United States have a college degree compared to 42.1 percent in Canada and 42.2 percent in the United Kingdom. Lower levels of education among Asian immigrants to the United Kingdom may partly be the result of the greater focus of immigration policy in the United Kingdom. In Canada, however, we would expect the point-based system of immigration to result in higher education levels among Asian immigrants than the United States. For every group except Koreans, Asian immigrants in the United States are more educated than those residing in Canada. Although there are many institutional, structural, and historical differences between the two countries that might be responsible, one possibility is that the higher returns to education in the United States result in a more selective immigrant pool. The researchers find that the returns to a college degree in business earnings are larger in the United States than in Canada. The returns to a college degree are also higher in the wage and salary sector in the United States than in Canada.
Is the vast army of the self-employed in low income countries a source of employment generation? De Mel, McKenzie, and Woodruff use data from surveys in Sri Lanka to compare the characteristics of own account workers (non-employers) with wage workers and with owners of larger firms. They use a rich set of measures of background, ability, and attitudes, including lottery experiments measuring risk attitudes. Consistent with ILO's views of the self employed (represented by Tokman), the researchers find that two-thirds to three-fourths of the own account workers have characteristics more like wage workers than larger firm owners. This suggests that the majority of the own account workers are unlikely to become employers. Using a two and a half year panel of enterprises, the researchers further show that the minority of own account workers who are more like larger firm owners are likelier to expand by adding paid employees. This analysis explains the low rates of growth of enterprises supported by microlending.
Using a unique census dataset on all industrial firms (with more than 5 million yuan in sales), Qian and Huang document the phenomenon of missing entrepreneurship in Shanghai. Here, entrepreneurship is defined as private, new entrants. Specifically, in terms of employment size and growth, the relative ranking of Shanghai has always been near the bottom in the country. All of the empirical findings exist against a backdrop of the presumably huge locational advantages of Shanghai: the substantial human capital, rapid GDP growth, and a long and stellar, but pre-communist history of entrepreneurship. The authors propose a hypothesis that Shanghai adopted a particularly rigorous version of an industrial policy model of economic development and that this industrial policy proclivity led to the atrophy of entrepreneurship in Shanghai.
Ardagna and Lusardi use a micro data set that collects information across individuals, countries, and time to investigate the determinants of entrepreneurial activity in 37 developed and developing nations. They focus both on individual characteristics and on countries' regulatory differences. They show that individual characteristics, such as gender, age, and status in the work force are important determinants of entrepreneurship, and they also highlight the relevance of social networks, self-assessed skills, and attitudes toward risk. Moreover, they find that regulation plays a critical role, particularly for those individuals who become entrepreneurs to pursue a business opportunity. The individual characteristics that are affected most by regulation are those measuring working status, social network, self-assessed skills, and attitudes toward risk.
Why do governments set up market entry barriers? To answer this question,Mullainathan and Schnabl study a reform of entry regulation in Lima, Peru. They show that the reform reduced time-to-license from 110 to 5 days and increased the annual number of new licenses from 2,000 to 9,000. They further show that the reform focused on reorganizing internal government processes, but did not change licensing requirements targeted at fixing market failures. Interviews with newly licensed firms show that the primary motivation for getting a license is to avoid fines and bribes. These researchers argue that this evidence is consistent with a bureaucracy that sets up market entry barriers to extract rents.
Brander, Egan, and Hellmann investigate the relative performance of enterprises backed by government-sponsored venture capitalists and private venture capitalists. While previous studies focus mainly on investor returns, this paper focuses on a broader set of public policy objectives, including value-creation, innovation, and competition. A number of novel data-collection methods, including web-crawlers, are used to assemble a near-comprehensive dataset of Canadian venture-capital backed enterprises. The results indicate that enterprises financed by government-sponsored venture capitalists underperform on a variety of criteria, including value-creation (as measure by the likelihood and size of IPOs and M&As), or innovation (as measured by patents). While the data does not allow for a definitive welfare analysis, the results cast some doubt on the desirability of government intervention in the venture capital market.
Djankov and his co-authors test two competing hypotheses on what makes an entrepreneur: nature -- attitude towards risk, IQ, and self-confidence; or nurture -- family background and social networks. To do so, they report the results of a new survey on entrepreneurship in Brazil, based on interviews of 400 entrepreneurs and 550 non-entrepreneurs of the same age, gender, education, and location in seven Brazilian cities. They find that family characteristics have the strongest influence on becoming an entrepreneur. In contrast, success as an entrepreneur is primarily determined by the individual's smartness and higher education in the family. Entrepreneurs are not more self-confident than non-entrepreneurs; and overconfidence is bad for business success.
These papers will be published by the University of Chicago Press in an NBER Conference Volume. They will also appear in "Books in Progress" on the NBER's website.
How do private returns to inventive activity change in developing countries when IPR regimes are substantially strengthened? Arora and his co-authors investigate this question by looking at the impact of patent reforms in India on India-based pharmaceutical companies. In a fundamental policy shift, India agreed to introduce product patents for pharmaceuticals when it signed the WTO TRIPS treaty in 1995. This policy came into effect through enabling legislation in 2000 and final implementation in 2005. The authors estimate the impact of this policy shift by using data on a panel of 315 Indian pharmaceutical firms drawn from the years 1990 to 2005. They find evidence of an increase in both R and D investment and measured inventive output that appears to be broadly coincident with patent reform. They also find that the private returns to R and D investment appear to be rising as a consequence of patent reform. Private returns to firms' investments are measured using a hedonic stock market valuation of tangible total assets (A) and intangible inventive assets (K), measured as the stock of R and D spending. The findings indicate an economically and statistically significant increase in private returns to inventive activity. However, this effect appears to be highly concentrated in the most technologically progressive Indian firms.
Berndt and his co-authors report early stage findings from a long-term research program that seeks to understand factors affecting the increasing globalization of biopharmaceutical clinical trials (BCTs) for new medicines, particularly into emerging economies. Previous literature dealing with the effects of intellectual property protection on innovation has been challenged by difficulties in quantifying both intellectual property and innovation. The relatively narrow focus here -- assessing impacts of several alternative measures of intellectual property protection on a country's level and AAGR of global share of clinical trial sites -- has the advantage of focusing on a specific type of investment in development (D), not justoverall R and D. Using data accessible from clinicaltrials.gov, the authors quantify and model the cumulative number of BCTs in the top 50 countries, and 2002-6 AAGRs in a country's share of BCT sites registered at clinicaltrials.gov. Although the United States, Western Europe, and Canada still dominate in terms of cumulative numbers of BCT sites, in general there has been a rapid growth in BCT numbers and shares in Eastern Europe, Latin America, and Asia, at the expense of Western Europe and North America. The researchers find that the elasticity of cumulative BCT sites with respect to GDP is about one, while the elasticity with respect to cost per patient is about -0.8; another important factor having a positive impact is the country's human capital index. Among three alternative measures of intellectual property protection developed by Park, the researchers find that only the broad measure encompassing 1990-2000 changes in patent coverage of pharmaceutical, chemical, and surgical tool and instrument products has a substantial and statistically significant positive impact. In terms of cross-country variations in AAGRs of trial sites, the results are largely consistent with emerging economies "catching up" with slower growing countries traditionally involved in clinical medicine. AAGRs are negatively correlated with GDP per capita, and to the cumulative number of first authors of articles reporting results from randomized clinical trials in major MedLine journals. The researchers also find that the 2000 level of the two broader measures of intellectual property protection have positive and significant impacts on growth rates of shares of BCT sites.
Thursby and Thursby examine data from a survey of 250 multinational firms regarding the location of recently established R and D facilities and the factors related to the location decisions. Their results show that R and D location decisions are quite complex and influenced by a variety of factors. In addition, while many of the facilities identified in the survey were located in emerging economies, a substantial portion of new facilities were in developed economies. The survey also asked firms to characterize whether the science conducted at these new facilities is "new" or "familiar" and whether the purpose of the science is for a "new" market for the firm or one that is "familiar" to the firm. Thursby and Thursby then analyze whether the type of science (new or familiar) and the purpose of the science (new or familiar markets) are functions of location characteristics; they also compare results for healthcare firms versus firms in other industries and find some striking differences. Healthcare firms conduct more cutting edge science than other firms do. Collaborating with universities, and the quality of healthcare personnel, is more important to healthcare firms for the type of science. Healthcare firms are more likely than others to conduct R and D for new markets. The likelihood of healthcare firms conducting R and D for new markets is positively related to ease of collaboration with universities and faculty with specialized expertise.
Fabrizio and Thomas examine innovations of new drugs in the global pharmaceutical industry. They contrast anticipated demand and historical technological expertise as determinants of the realized pattern of innovations at the country level. They further contrast local versus foreign determinants of innovation. They find that the pattern of demand is as important as technological expertise in determining the pattern of innovation in this industry. They also find that innovation is a locally determined phenomenon, with very little evidence of positive cross-country knowledge spillovers.
Over the past 15 years, academic medical centers have ceased to be the primary locus of industry-sponsored clinical trial activity. Instead, clinical trials have increasingly been conducted in private practices and for-profit, dedicated study sites. Azoulay and Fishman examine the underlying causes of this startling evolution. On the demand side, the greater availability of non-academic investigators has enabled pharmaceutical firms to better match physicians' skills with specific projects. On the supply side, the authors argue that the growth of managed care health insurance has contributed to a rise in the number of non-academic physicians performing clinical research. They find evidence consistent with these claims using a unique dataset containing information about 85,919 site contracts for 7,735 clinical trials between 1991 and 2003. Furthermore, they examine the gap in prevailing prices for comparable procedures conducted for clinical trials versus conventional medical care, and conclude that the effect of managed care on entry is consistent with non-academic physicians "inducing demand" so as to resist downward pressures on their income.
Over the past few decades, large pharmaceutical firms have merged at a high rate. Numerous motives appear to be have encouraged this M&A activity, including the desire to refill emptying pipelines of new drugs, to achieve economies of scale and scope in increasingly costly R and D projects; and to gain access to marketing, distribution, and/or drug testing/approval networks and expertise. Regardless of the motives driving these mergers, managers of the newly combined firm face difficult questions regarding whether and how to re-organize its various R and D organizations in the post-merger period. Addressing this question involves a dilemma: whether to close laboratories and centralize R and D geographically (and possible organizationally as well) in order to capture economies of scale and scope within or across R and D projects, or to retain and possibly expand a dispersed structure of R and D organizations in order to continue to gain access to a variety of loci of knowledge in fundamental life sciences as well as drug development. Furman and his co-authors examine data that reflect the outcomes of the ways in which large pharmaceutical firms have been managing this trade-off. They study mergers and their attendant effects on the location of pharmaceutical research between 1984 and 1999. Their primary empirical approach takes advantage of NBER patent data and uses difference-in-differences techniques to examine the impact of mergers on the location of drug discovery research. They supplement their large-scale empirical analysis with brief qualitative accounts of the lab dispersion decisions of merged firms.
Delgado and her co-authors examine changes in the patterns of global trade in biopharmaceuticals before and after the implementation of the WTO TRIPS agreement on intellectual property. The purpose of their analysis is to explore the evidence consistent with the objective of TRIPS, to promote "the transfer and dissemination of technology" particularly from advanced to least-developed countries. The evidence suggests that, during the period of TRIPS implementation between 1994 and 2005, global trade in biopharmaceuticals and other products dependent on intellectual property increased relative to sectors unaffected by TRIPS. But, the results also suggest that trade in technology-intensive products has not grown dramatically between developed and less developed countries. To explore alternative explanations for the patterns, the statistical analysis is supplemented with brief case analyses of the impact of TRIPS in South Africa, the United Kingdom, India, and Brazil.
As the value chain of the pharmaceutical industry disaggregates, upstream discovery is increasingly carried out by small research-specialized firms while downstream development, testing and marketing is conducted by global pharmaceutical firms. Licensing plays an important role in this emerging division of labor. Alcacer and his co-authors theorize that, similar to markets for upstream inputs such as scientific knowledge, proximity also may matter for licensing, which they conceptualize as downstream end markets for small biotechnology firms. They examine whether co-location affects the likelihood of vertical licensing transactions between biotechnology firms and global pharmaceutical firms. Discussions with industry executives indicate that large firms search globally for in-licensing opportunities and that licensing transactions should not be sensitive to the geographic locations of the transacting parties. However, an analysis of compounds developed by small biotechnology firms licensed to global pharmaceutical firms suggests that licensing transactions are more likely to occur between firms located in the same geographic area. The results point to the possibility that licensing markets are sensitive to the proximity of the partners, and that despite global search processes by multinationals in the pharmaceutical industry, licensing markets are localized.
Koenig and MacGarvie examine the relationship bewteen cross-country differences in drug price regulation and the location of biopharmaceutical Foreign Direct Investment (FDI) in Europe. They use a theoretically-grounded location-choice model and data on 294 investments initiated in 27 European countries between 2002 and 2006 to test the hypothesis that biopharmaceutical companies are less likely to locate new investments in countries with more stringent price regulation.
To most tax policy analysts and academics, the term "tax expenditure" means special provisions of the tax system that result in reduced tax liability for certain subsets of taxpayers. Moreover, for many in the tax policy community, the term suggests tax breaks for limited constituencies that result in a narrow tax base and higher marginal tax rates. Others are blunter: tax expenditures are loopholes that need to be closed. Altshuler and Dietz leave it to other authors to examine the legitimacy from tax policy or economic efficiency perspectives of tax expenditure provisions. They instead examine the measurement of tax expenditures and offer recommendations aimed at improving their value to analysts and policymakers. They use calculations from NBER's TAXSIM to illustrate some of the problems with the current methodology for estimating tax expenditures. Unlike most previous work on the topic, theirs focuses on how features of the current tax system that were not in place when Stanley Surrey introduced the expenditure concept in 1967, such as the alternative minimum tax (AMT) and sunset rules, complicate and compromise the value of information provided by the tax expenditure budget. Their recommendations for reform include presenting revenue estimates for major tax expenditures, reporting some negative tax expenditures, and establishing an appendix for tax expenditure estimates of permanent versions of expiring provisions and AMT interaction effects, among others. These proposed changes to existing tax expenditure estimation process will increase the value of tax expenditure reporting for academics, policymakers, and others in the tax analysis community.
Carroll and his co-authors explore the implications of evaluating income tax expenditures under a consumption tax. First, they examine the conceptual differences between income and consumptions taxes. Next, they use an X-tax prototype of a consumption tax to gauge the sensitivity of the estimates to the two baselines: current law income tax and the X-tax. The estimated capital income preferences are vastly different under the two regimes.
Guthrie and Hines consider the impact of the tax treatment of U.S. military contractors on government procurement contracts. Prior to passage of the Tax Reform Act of 1986, taxpayers were permitted to use the completed contract method of accounting to defer taxation of profits earned on long-term contracts. The Tax Reform Act and subsequent legislation passed in 1987 required that at least 70 percent of the profits earned on long-term contracts be taxed as accrued, thereby significantly reducing the tax benefits associated with long-term contracting. Comparing contracts that were ineligible for the tax benefits associated with long-term contracting with those that were eligible, it appears that between 1981 and 1989 the duration of U.S. Department of Defense contracts shortened by an average of between one and 3.5 months, or somewhere between 6 and 29 percent of average contract length. This pattern suggests that the tax benefits associated with long-term contracts promoted artificial contract lengthening prior to passage of the 1986 Act. The evidence is consistent with a behavioral model in which the Department of Defense ignores the federal income tax consequences of its procurement actions, thereby indirectly rewarding contractors who are able to benefit from tax expenditures of various types.
The tax exclusion of employer-sponsored insurance from taxable income costs the United States almost $250 billion per year in foregone revenue. In his paper, Gruber discusses the rationales for and against this tax subsidy, present estimates of its magnitudes and distributional consequences, and show the impacts of a variety of approaches to reform.
Ackerman and Auten examine tax expenditure for noncash charitable giving. They begin with an analysis of issues that arise specifically with donations of property rather than cash. These concerns include valuation of the donated property, potential differences between the value for donor and donee, design and enforcement of appropriate tax rules, the relative responsiveness of noncash donations to the deduction incentive, and the potential effects on charities that accept these gifts. Using a special study of tax return data from 2003-5, the researchers show the deductions claimed for different types of property. Donations of corporate stock accounts for the largest dollar amount, but two-thirds of those deducting over $500 in non-cash items donate clothing, and these deductions actually exceeded real estate donations by 2005. The authors then explore vehicle donations in more detail. Using a unique dataset that combines tax return information on vehicle donations with sales information from on-line auctions, they show that prior to the law change in 2005, the implied quality of the vehicles donated by taxpayers exceeded plausibility. They conclude by summarizing recent proposals for reform, such as increasing reporting requirements, tightening valuation and appraisal standards, and limiting or eliminating the additional tax benefit from donating appreciated property.
Bakija and Heim estimate the elasticity of charitable giving with respect to tax prices and aftertax incomes using a panel of nearly 500,000 disproportionately high-income tax returns spanning the years 1979 through 2005. They improve upon the previous literature by using state tax variation to help identify their model while controlling for unobserved heterogeneity, allowing people at different income levels to have different degrees responsiveness to taxation, and carefully dealing with the fact that, because of lags between proposal, enactment, and implementation of tax reforms, near-future changes in taxes are generally predictable in advance. To address the omitted variable bias that would otherwise arise from failing to control for unobservable expectations of future prices and future incomes, they use predictable changes in future federal and state marginal tax rates and tax liabilities, arising from their pre-announced and phased-in nature, as instruments for future changes in prices and income. In models where identification of price effects comes largely from different time paths of marginal income tax rates across different states and across people at different income levels within the same income class, there is robust evidence of a modest but statistically significant elasticity of charitable giving with respect to persistent changes in tax-price among people with incomes in excess of $100,000, generally in the range of -0.5 to -0.8. Evidence on the persistent price elasticity for lower-income people is weaker and more mixed. Despite a large spike in giving among very-high income people in apparent anticipation of lower future tax benefits from charitable giving arising from enactment of the Tax Reform Act of 1986, there is surprisingly little evidence that people re-time their giving in response to anticipated differences between current and future tax savings from doing so over the sample period as a whole and across income groups. Several results here are consistent with the permanent income hypothesis. The estimates suggest that expenditures on charitable giving respond strongly to persistent changes in income, while responding very little to transitory fluctuations in income. Moreover, there is strong evidence that people will increase their charitable giving now in response to a predictable reduction in future tax liability arising from tax reform.
Poterba and Sinai examine the effects of the mortgage interest deduction, the property tax deduction, and the absence of taxation on imputed rent on the effective cost of housing services, federal income tax revenues, and the distribution of tax liabilities. They consider how several changes in these tax provisions would affect taxpayer incentives, and assess the potential revenue consequences of each. Their analysis recognizes that changing tax provisions, such as the mortgage interest deduction, would induce changes in homeowner behavior, both with respect to housing finance and the quantity of housing demanded. These changes can affect estimates of the revenue cost of housing-related tax expenditures. With regard to the mortgage interest deduction, for example, they estimate that without any behavioral response, income tax revenues would have been $69.8 billion higher in 2003. Allowing for portfolio adjustments in the financing of homes reduces this estimate to $63.4 billion, and the value could be lower still if they made alternative assumptions about financial changes. The results point generally to the importance of recognizing behavioral responses to tax incentives when calculating the revenue costs of tax expenditures.
The Low Income Housing Tax Credit (LIHTC) represents a novel tax expenditure program that employs "investable" tax credits to spur production of low-income rental housing. While it has grown into the largest source of new affordable housing in the United States and is now being replicated in other programs, its curious structure has also drawn skepticism and calls for its repeal. Desai and his co-authors outline a conceptual framework for exploring the conditions under which investable tax credits may be the most effective mechanism for delivering a production subsidy and discuss the desirability of employing investable tax credits in other policy domains. Estimates of tax expenditures under this program are provided and efficiency costs, distributional issues, and the likely effects of reforms to tax provisions such as the AMT are considered.
In the United States, the primary means of providing cash assistance to lower-income families with children is now the federal income tax system. A series of tax acts beginning with the 1986 Tax Reform Act running parallel to the erosion of the traditional welfare system have increased assistance to the working poor through expansions of the Earned Income Tax Credit (EITC). In 2007, an estimated 22 million families are estimated to benefit from the EITC, at a total cost to the federal government of more than 43 billion dollars (U.S. Office of Management and Budget, 2008). Eissa and Hoynes examine the distributional and behavioral effects of the Earned Income Tax Credit. They first review the current design of the credit and its growth over time, and then examine the evidence on individual behavioral responses. They argue that EITC expansions show that real responses to taxes are important. As for distribution, the tax data show that the current credit transfers cash to families who are well above the median of the cash-income distribution. Eissa and Hoynes then examine the cost, coverage, and change in the distribution of benefits under hypothetical reforms of the credit. Finally, using survey data from the Current Population Survey, they calculate the efficiency effects of marginal changes in EITC parameters. Targeting the EITC to lower-income families by raising the phase-out rate would generate a welfare loss for single mothers, primarily because of the disincentive to enter the labor market.
Federal deductibility for state and local taxes constitutes one of the largest tax expenditures in the federal budget and provides a significant source of federal support to state and local governments. Deductibility was restricted in the Tax Reform Act of 1986 by removing the deduction for general sales taxes. More recently, the President's Advisory Panel on Federal Tax Reform recommended eliminating the deduction altogether as one of several revenue-raising initiatives to finance comprehensive tax reform. Feldstein and Metcalf (1987) argued that estimates of the revenue gain from eliminating deductibility were too high, as they did not take into account a likely shift away from once-deductible taxes to non-deductible taxes and fees in the absence of deductibility. Many of these latter taxes and fees are paid by businesses. As business costs rise, federal business tax collections would fall, offsetting some of the gains of ending deductibility. Feldstein and Metcalf also found that ending deductibility would have little if any impact on state and local spending itself. Using a large panel of data on state and local governments, Metcalf here revisits this issue and finds that the Feldstein-Metcalf results are robust to adding more years of analysis. He then carries out a number of distributional analyses considering both variation across income and across states of the subsidy from deductibility. In addition, he considers three counterfactuals for 2004 a tax system without the Bush tax cuts for 2001 and 2003, a tax system without the 2004 AMT patch, and a tax system without the AMT to see how the benefits of deductibility are affected by these changes in the tax law.
Poterba and his co-author explore how estimates of the revenue cost of exempting interest payments by state and local governments from the federal income tax are affected by alternative assumptions about the portfolio behavior of individual investors. Most tax expenditure estimates assume that current holders of tax-exempt bonds would replace their holdings with taxable bonds if the tax expenditure were eliminated. The authors consider a number of alternative possible portfolio responses and examine how they would affect estimates of the aggregate revenue cost of tax exemption as well as the distribution of tax burdens. Because taxable bonds are among the most heavily taxed assets, the assumption that investors holding tax-exempt bonds switch to taxable bonds yields a larger estimate of the revenue cost of tax exemption than alternative portfolio response assumptions. Using household-level data from the 2004 Survey of Consumer Finances, the researchers estimate that the revenue cost of tax exemption under the taxable bond substitution hypothesis is $14.2 billion, compared with $10.1 billion if they assume that corporate stock replaces tax-exempt bonds and $7.9 billion if they assume that investors distribute their tax-exempt bond holdings in proportion to their current portfolio holdings of all asset classes. They also explore the revenue effects of other policy alternatives to full elimination, such as capping the dollar amount of tax-exempt interest per tax return or limiting tax-exempt interest as a fraction of AGI.
These papers will be published by the University of Chicago Press in an NBER Conference Volume. They will also appear in "Books in Progress" on the NBER's website.
Hotchkiss and her co-authors examine whether, and how, leveraged buyouts from the most recent wave of public-to-private transactions create value. For a sample of 192 buyouts completed between 1990 and 2006, they show that these deals are somewhat more conservatively priced and lower levered than their predecessors from the 1980s. For the subsample of deals with post-buyout data available, median market adjusted returns to pre- and post-buyout capital invested are 78 percent and 36 percent, respectively. Gains in operating performance, however, are either comparable to or slightly above those observed for benchmark firms. The researchers find that cash flow gains are greater for firms with larger increases in leverage, and when the CEO has been replaced at the time of the buyout. Finally, they examine the relative contribution of potential determinants of returns; in addition to gains in operating performance, returns are driven by increases in industry valuation multiples and are greater for deals involving more than one private equity firm. Overall, these results provide insights into how transactions from the most recent wave of leveraged buyouts create value.
Axelson and his co-authors provide an empirical analysis of the financial structure of large recent buyouts. They collect detailed information on the financings of 153 large buyouts (averaging over $1 billion in enterprise value). They document the manner in which these important transactions are financed. Buyout leverage is cross-sectionally unrelated to the leverage of matched public firms, and is largely driven by other factors than what explains leverage in public firms. In particular, the economy-wide cost of borrowing seems to drive leverage. Prices paid in buyouts are related to the prices observed for matched firms in the public market, but are also strongly affected by the economy-wide cost of borrowing. These results are consistent with a view in which the availability of financing impacts booms and busts in the private equity market.
Ljungqvist and his co-authors analyze the determinants of buyout funds' investment decisions. In a model in which the supply of capital is "sticky" in the short run, the researchers link the timing of funds' investment decisions, their risk-taking behavior, and the returns they subsequently earn on their buyouts to changes in the demand for private equity, conditions in the credit market, and funds' ability to influence their perceived talent in the market. Using a proprietary dataset of 207 buyout funds that invested in 2,274 buyout targets over the last two decades, the researchers then investigate the implications of their model. Their dataset contains precisely dated cash inflows and outflows in every portfolio company, links every buyout target to an identifiable buyout fund, and is free from reporting and survivor biases. Thus, they are able to characterize every buyout fund's precise investment choices. Their empirical findings are consistent with the model. First, established funds accelerate their investment flows and earn higher returns when investment opportunities improve, competition for deal flow eases, and credit market conditions loosen. Second, the investment behavior of first-time funds is less sensitive to market conditions. Third, younger funds invest in riskier buyouts, in an effort to establish a track record. Fourth, following periods of good performance, funds become more conservative, and this effect is stronger for younger funds.
Groh and Gottschalg measure the risk-adjusted performance of U.S. buyouts. They draw on a unique and proprietary set of data on 133 U.S. buyouts between 1984 and 2004. For each of them, they determine a public market equivalent that matches it with respect to its timing and its systematic risk. After a correction for selection bias in the data, the regression of the buyout internal rates of return on the internal rates of return of the mimicking portfolio yields a positive and statistically significant alpha. The sensitivity analyses highlight the necessity of a comprehensive risk-adjustment that considers both operating risk and leverage risk for an accurate assessment of buyout performance. This finding is particularly important as existing literature on that topic tends to rely on performance measures without a proper risk-adjustment.
Driessen and his co-authors develop a new methodology to assess the abnormal performance and risk exposure of a non-traded asset from a cross-section of cash flow data. They apply this method to a sample of 958 mature private equity funds spanning 24 years. In contrast to existing work, their methodology mainly uses actual cash flow data and not intermediary self-reported net asset values. They find a beta for venture capital funds above 3 and a beta for buyout funds below 1. Venture capital funds have significantly negative abnormal performance while the abnormal performance of buyout funds is close to zero. Larger funds have higher returns because of higher risk exposures, not higher alphas. The authors also show that Net Asset Values overstate fund market values for the subset of mature and inactive funds.
Katz explores the change in earnings management and conservatism as firms backed by private equity (PE) sponsors make the transition between private and public ownership. Using a unique sample of U.S. firms, she compares a private phase, in which firm equity is privately held while firm debt is publicly held, to a public phase, in which firm equity is also publicly held. In addition, she separately analyzes PE-backed firms and non-PE-backed firms within the private phase. Katz finds that during the public phase, PE-backed firms engage in greater upward earnings management in order to avoid small earnings decreases, and they recognize losses in a more timely manner than during the private phase. Furthermore, she finds that PE-backed private firms engage less in upward earnings management, and recognize losses more promptly, than do non-PE-backed private firms. These results are robust for various measures and controls, and are not affected by factors such as endogenous listing status and PE financing choices.
The impact of private equity on employment outcomes arouses considerable controversy. Critics claim that private equity buyouts bring huge job losses, while recent research sponsored by private equity associations and others claims big positive employment effects. To address the issue, Davis and his co-authors construct and analyze a new dataset that overcomes many of the limitations in previous research. They examine U.S. private equity transactions from 1980 to 2000, following 5,000 target firms and 300,000 establishments before and after acquisition by private equity groups. They compare employment outcomes at target firms and their establishments to controls that do not have private equity ties and that are similar in terms of industry, size, and age. The key findings are: 1) Employment declines more rapidly at target establishments than at controls in the wake of private equity buyouts. 2) Target establishments create roughly as many new jobs as control establishments post-buyout, but they destroy old jobs at a faster pace. 3) Employment also grows more slowly at target establishments than at controls in the two years preceding the buyout transaction. 4) The target firms acquired by private equity groups create more new jobs at new facilities than control firms. Combining this greater job creation response at new establishments with the greater job loss at existing establishments, the researchers find little difference in net job growth between controls and target firms in the wake of private equity buyouts. Taken together, these results suggest that private equity groups act as a catalyst for creative destruction in the labor market.
Using a detailed dataset with assessments of CEO candidates for companies involved in private equity (PE) transactions, including both buyout (LBO) and venture capital (VC) deals, Kaplan and his co-authors study how CEOs' characteristics and abilities relate to hiring decisions, PE investment decisions, and subsequent performance. The candidates are assessed on more than 40 individual characteristics in seven general areas leadership, personal, intellectual, motivational, interpersonal, technical, and specific. In general, all characteristics and abilities are found to be highly correlated. For both LBO and VC firms, outside CEO candidates are more highly rated than incumbents. Both LBO and VC firms are more likely to hire and invest in more highly rated and talented CEOs, and the investors also value "soft" or team-related skills in the hiring decisions. However, these skills are not necessarily associated with greater success. For LBO deals in particular, "hard" abilities and execution skills predict success. Finally, the researchers find that incumbents are no more likely to succeed than outside CEOs, holding observable talent and ability constant.
Stuart and Yim examine the propensity for U.S. public companies to become targets for private equity-backed, take-private transactions. They consider the characteristics of 483 private equity-backed deals in the 2000-7 period relative to public companies and find that, in addition to the financial drivers studied in previous works, board characteristics and director networks are associated with deal generation. They further find that a company with a director who has had LBO experience through prior board service is about 40 percent more likely to receive a private equity offer, and that the strength of this effect varies with the influence of the director and the quality of the prior LBO experience. This effect is robust to the most likely alternative explanations and supports the idea that directors and social networks play an influential role in change-of-control transactions.
Leslie and Oyer analyze the differences between companies owned by private equity (PE) investors and similar public companies. They document that PE owned companies use much stronger incentives for their top executives and have substantially higher debt levels. However, there is little evidence that PE owned firms outperform public firms in profitability or operational efficiency. The researchers also show that the compensation and debt differences between PE owned companies and public companies disappear over a very short period (one to two years) after the PE owned firm goes public. Their results raise questions about whether and how PE firms and the incentives they put in place create value.