The NBER Reporter Spring 2004: Conferences
Fifth Annual Conference in India
Garleanu and Zwiebel analyze the design and renegotiation of covenants in debt contracts as a particular example of the contractual assignment of property rights under asymmetric information. In particular, they consider a setting in which future firm investments are efficient in some states, but also involve a transfer from the lender(s) to shareholders. While there is symmetric information regarding investment efficiency, managers are better informed about any potential transfer than the lender. The lender can learn this information, but at a cost. In this setting, the authors show that the simple adverse selection problem leads to the allocation of greater ex ante decision rights to the uninformed party than would follow under symmetric (in particular, full) information. Consequently, ex-post renegotiation in turn is biased towards the uninformed party giving up these excessive rights. In many settings, this result yields the opposite implication from standard Property Rights results on contracting under incomplete contracts and ex-ante investments, whereby rights should be allocated to minimize inefficiencies attributable to distortions in ex-ante investments. Indeed, for debt contracts as well as other settings, the uninformed party, who receives strong decision rights in this setting, is likely to have few significant ex-ante investments to undertake relative to the informed party.
B>Wernerfelt proposes a research program to compare game forms in terms of their ability to govern ex post adjustments to ex ante contracts. The comparisons can be based on direct implementation costs or on the extent to which desirable adjustments are not implemented. In several examples of the program, he compares three game forms: negotiation over each adjustment; ex ante price lists; and implicit contracts leaving the stipulation of adjustments to one player. If the latter game form is defined as an employment relationship, then the theory of the firm becomes a special case of the program. Wernerfelt starts with a discussion of the nature and magnitude of adjustment costs, then follows with an exposition of four examples. He then discusses the role of asset ownership, reviews some empirical evidence, and looks at broader implications.
The widespread use of strategic alliances between pharmaceutical and biotechnology companies is puzzling, since it is hard to contract on the exact nature of the research activities. A major concern of pharmaceutical companies entering strategic alliances is that the biotechnology firm will use the pharmaceutical company’s funds to subsidize other projects or substitute one project for another. Using a new dataset on 584 biotechnology strategic alliance contracts, Lerner and Malmendier find that the parties respond to this contracting problem by assigning the unconditional right to terminate the alliance, including the reversion of intellectual property rights, to the pharmaceutical company. The authors develop a model based on the property-rights theory of the firm that allows for biotechnology firm researchers to pursue multiple tasks. They show that it is optimal for the pharmaceutical company to obtain the right to terminate the alliance and to receive the property rights to the terminated project when research activities are non-contractible. This right will induce the biotechnology firm researchers not to deviate from the proposed research activities. The contract prevents opportunistic exercise of this termination right by specifying payments triggered by the termination of the agreement. Testing the model empirically, the authors find that the assignment of termination and product reversion rights to the financing firm occurs in contractually difficult environments in which the parties are unlikely to be able to specify the lead product candidate. They use empirical tests to distinguish the property-rights explanation from alternative stories, based on uncertainty and asymmetric information about the project quality or research abilities.
Lazear proposes that entrepreneurship consists of team-building and assembling resources. As such, entrepreneurs must be jacks-of-all-trades who need not excel in any one skill, but are competent in many. A model of the choice to become an entrepreneur, which he presents, primarily implies that individuals with balanced skills are more likely than others to become entrepreneurs. Those who want to start businesses acquire their general backgrounds through a varied course curriculum and by taking on a broad range of roles when they enter the labor force. Using a dataset of Stanford alumni, Lazear tests the predictions and finds that they hold. Entrepreneurs are not technical innovators. By far the most important prediction of entrepreneurship is having a varied work background.
Entrepreneurs often get their ideas from working as employees in established firms. However, employees with ideas also can become intrapreneurs, or even managers of corporate spin-offs. Hellman shows how innovation and entrepreneurship are influenced by company policies towards employees. Using a multi-task incentives model, he identifies a trade-off between focusing employees on their assigned tasks and encouraging their exploration of new ideas. He shows how the rate of innovation, and the organizational structure of new ventures (start-ups, spin-offs, internal ventures), depend on factors such as the entrepreneurial environment and the allocation of intellectual property rights.
Antras and Helpman present a North-South model of international trade in which differentiated products are developed in the North. Sectors are populated by final-good producers who differ in productivity levels. Based on productivity and sectoral characteristics, firms decide whether to integrate into the production of intermediate inputs or outsource them. In either case they have to decide from which country to source the inputs. Final-good producers and their suppliers must make relationship-specific investments, both in an integrated firm and in an arm’s-length relationship. The authors describe an equilibrium in which firms with different productivity levels choose different ownership structures and supplier locations, that is, they choose different organizational forms. Then they study the effects of within-sectoral heterogeneity and variations in industry characteristics on the relative prevalence of these organizational forms. The analysis sheds light on the structure of foreign trade within and across industries.
Acemoglu, Aghion, Griffith, and Zilibotti investigate the determinants of vertical integration and confront some of the predictions of the leading approach to the internal organization of the firm with data from the U.K. manufacturing sector. Consistent with the theory, the authors show that an upstream and downstream activity pair are more likely to be vertically integrated when the downstream (the producer) is more technology intensive and the upstream (the supplier) is less technology intensive. Also consistent with the theory, the magnitude of both effects are substantially amplified when the upstream inputs are an important fraction of the total costs of the downstream producer. These results are generally robust and hold with a variety of alternative measures of technology intensity, with alternative estimation strategies, and with or without controlling for a number of firm and industry-level characteristics.
How does firm investment change with cash flow? Bertrand and Mullainathan examine this question for auctions of oil and gas leases because detailed data on specific investment projects are publicly available in this context. All bids, including losing ones, as well as the eventual outcome of the leases can be measured. The authors find that as cash flow rises, firms do spend more on purchasing leases. Interestingly, though, they do not buy more or bigger leases; instead, they simply pay more for each lease. This effect is strongest as firms approach the end of their fiscal year. Leases bought when cash flow is high are not more productive; in face, they are often less productive. In short, when cash flow is high, bidders appear to over-pay for less productive leases without expanding the scale of operations. These results are most consistent with a free cash-flow view of investment in which managers use cash flow to simplify their job (or live a “quiet life”) rather than empire build. The authors also find that the productivity effects are strongest earlier in their sample, consistent with the view that governance in this industry has improved over time.
Bertrand, Schoar, and Thesmar investigate the effects of banking deregulation on changes in banks’ lending behavior and the ensuing incentives for firms to improve operations. Most importantly, they analyze the implication of these changes on exit and entry decisions of firms and overall product market structure in the non-financial sectors. They use the deregulation of the French banking industry in 1985 as an economy-wide shock to the banking sector that affected all industries, but in particular those that relied most heavily on external finance and bank loans. The deregulation eliminated government interference in lending decisions, allowed French banks to compete more freely against each other in the credit market and did away with implicit and explicit government subsidies for most bank loans. Post-deregulation, banks seem to tie their lending decisions more closely to firm performance. Low quality firms that suffer negative shocks do not receive large increases in bank credit anymore. Instead, these firms display a much higher propensity to undertake restructuring measures post-reform, for example to reduce wages and outsource production. The authors also observe a strong increase in performance mean reversion post-1985, especially for firms that were hit by negative shocks. Moreover, they find that poorly performing firms experience a steeper increase in the cost of capital after the reforms than good firms. All these results are particularly strong for firms in more bank-dependent industries. On the product market side, the authors observe a strong increase in asset reallocation in more-bank-dependent industries, mostly coming from higher entry and exit rates in these sectors. They also find an increase in allocative efficiency across firms in these sectors, as well as a decline in concentration ratios.
Does the level of integration of a firm affect the quality of information available to its top decisionmakers responsible for allocating resources? Motivated by the pervasiveness of specific knowledge in large multi-division firms, Ozbas develops a model of internal competition for corporate resources among specialist managers and shows that: 1) managers of integrated firms exaggerate the payoffs of their projects to obtain resources despite potentially adverse career consequences; and 2) the exaggeration problem worsens with increased integration and reduces the allocative efficiency of an integrated firm. Control rights based on asset ownership enable the firm to set “the rules of the game” and to improve managerial behavior through organizational processes such as rigid capital budgets, job rotation, centralization, and hierarchies.
The popular press and scholarly studies have noted a number of trends in corporate governance. Hermalin addresses, from a theoretical perspective, whether these trends are linked and if so, how? He finds that a trend toward greater board diligence will lead, sometimes through subtle or indirect mechanisms, to trends toward more external candidates becoming CEO, shorter tenures for CEOs, more effort/less perquisite consumption by CEOs (even though such behavior is not directly monitored), and greater CEO compensation. Also, under plausible conditions, externally hired CEOs should have shorter tenures on average than internally hired CEOs.
Murphy and Zabojnik reconcile two pronounced trends in U.S. corporate governance: the increase in pay levels for top executives, and the increasing prevalence of appointing CEOs through external hiring rather than internal promotions. They propose that these trends reflect a shift in the relative importance of “managerial ability” (transferable across companies) and “firm-specific capital” (valuable only within the organization). They show that if the supply of workers in the corporate sector is relatively elastic, an increase in the relative importance of managerial ability leads to fewer promotions, more external hires, and an increase in equilibrium average wages for CEOs. They test their model using CEO pay and turnover data from 1970 to 2000. They show that CEO compensation is higher for CEOs hired from outside their firm, and for CEOs in industries where outside hiring is prevalent.
Internal language, or “codes,” constitute an important part of the shared tacit understanding jointly held by members of a group or organization. This shared understanding is often an integral part of a group’s culture and reflects important elements of the culture. Using a paradigm developed by Weber and Camerer to study such codes — as a simple metaphor for group or organizational culture — in the laboratory, Camerer, Weber, and Rick explore the interaction between internal language and firm structure in determining outcomes related to firm performance. Subjects in their experiments perform a repeated task in which one subject (the “manager”) has to get a group of other subjects (the “employees”) to identify a series of pictures by describing only the content of the pictures. To perform this task efficiently, the group must develop a set of codes. In the experiments, the authors use a very simple treatment variable: they vary the degree of centralization or hierarchy in their laboratory “firms” either by fixing the role of manager on one subject or by rotating it among all subjects. They then examine the ability of each type of laboratory firm to deal with problems similar to those encountered by real-world firms. In a first experiment, the authors examine performance in a repeated but static setting (the same group performing the task for 20 rounds) and in a changing environment (introducing new pictures and new members). While most measures of performance favor the centralized, hierarchical structure — in which the same subject is always the manager — they find that the ability of the group to assimilate new entrants is greater under the decentralized, egalitarian structure. In a second experiment, they test the extent to which the two kinds of structures produce differences in group solidarity by having subjects fill out questionnaires and play a public good game. They find that the decentralized, egalitarian structure produces more favorable attitudes towards the group and greater contributions to the public good.
Cremer, Garicano, and Prat study the determination of specialized codes under bounded rationality, and its implications for organization. Agents may decrease communication costs by designing codes that fit their own environment, for example by using more precise words for more frequent events. Bounded rationality imposes sharply decreasing returns to scope, since when similarly skilled agents in different services must communicate with one another they must share common codes, which in turn degrades communication within each service. Thus the decision of whether to segregate services or integrate them trades off the synergies that result from better coordination between services against the loss attributable to the need for a common, more vague, code than the one that would optimize communication within services. Alternatively, more skilled ‘translators’ may be used to allow separate services to appropriate the synergies while keeping their own codes. A decrease in diagnosis costs leads to increasing integration among services and to the substitution of common codes for hierarchies, as common codes allow for the direct interaction among agents in different services. When adoption decisions are decentralized and non-contractible, the common code will be inefficiently biased towards the needs of early adapters and there will be too little commonality of codes.
Fifth Annual Conference in India
On January 17-20, 2004 the NBER and India's National Council for Applied Economic Research (NCAER) again brought together a group of NBER economists and about two dozen economists from Indian universities, research institutions, and government departments for their fifth annual conference in India. Mihir A. Desai and Martin S. Feldstein, NBER and Harvard University, and Raghuram G. Rajan, NBER and University of Chicago, organized the conference jointly with Suman Bery and Shashanka Bide of NCAER.
The U.S. participants were: Jagdish Bhagwati, NBER Director and Columbia University; Marianne Baxter and Robert G. King, NBER and Boston University; Michael D. Bordo, NBER and Rutgers University; Mihir A. Desai, Martin S. Feldstein, and Benjamin M. Friedman, NBER and Harvard University; Esther Duflo, NBER and MIT; Karen N. Horn, NBER Director; Anne O. Krueger, on leave from the NBER at the IMF; Karthik Muralidharan, Harvard University; Edward P. Lazear and Kathryn Shaw, NBER and Stanford University; Richard Portes, NBER and Columbia University; and Helene Rey, NBER and Princeton University.
After introductory remarks about the U.S. and Indian economies by NBER President Feldstein and Bimal Jalan of NCAER, the participants discussed: monetary and fiscal policy; financial sector reforms; economic reforms; infrastructure and regulation; economic recovery; and growth and productivity.
A summary of the conference discussion will be available on the NBER web site here.