STATEMENT OF RESEARCH ACTIVITIES Gonçalo Pacheco de Almeida | HEC Paris

INTRODUCTION Most accepted definitions of business strategy are variations on one idea: that strategy is a set of actions to achieve the long-term goals of a firm. If strategy’s success is measured by its future outcomes, time – and the dynamic effects of firm choices over time – must be an explicit component of strategy research. My work on strategy dynamics shows that endogenizing the time dimension of firms’ decisions in simple theoretical and empirical models can address several outstanding puzzles in the strategy literature. In particular, I study three key types of problems in strategy dynamics: (1) compression, or the time acceleration of firms’ activities, processes and investments, (2) timing, or the sequencing and spacing in time of firms’ actions (i.e., not how fast, but when events occur), and (3) punctuality, or time coordination among agents and between organizations. I present these research streams in approximate chronological order below. My work on themes (1) and (2) is closely interconnected and, thus, is discussed together in the same section. RESEARCH STREAMS (1) AND (2): COMPRESSION AND TIMING My research on time compression and timing is motivated by a simple tradeoff: that accelerating firms’ pursuit of any strategic goal (competitive advantage, industry leadership, or innovation) typically comes at the expense of extreme investment inefficiencies. Accelerating firms’ activities may increase the revenue yield of new products and technologies because they are commercialized earlier. However, compressing time also substantially raises costs – at an increasing rate. These adjustment costs are known as time compression diseconomies.i Relatively little work in strategy and economics has investigated this time compression tradeoff. There are three reasons for this. Theoretical models often assume that (A) firm investments are instantaneous or that (B) firms do not decide on the level of acceleration of their activities, for analytical tractability. Empirical research usually (C) neglects the time-consuming nature of strategy because time is often an unobservable construct. The problem with these oversimplifying assumptions and empirical omission is the hazard of producing dynamically inconsistent normative results. Figure 1 illustrates this point. It is the existence of the time compression tradeoff that makes acceleration a strategic decision: an optimal level of acceleration in firms’ activities usually exists. Ignoring it leads to biased inferences about strategy fundamentals such as firms’ speed of innovation and imitation, the rate of technology diffusion in an industry, and the length of sustainable competitive advantage. My research relaxes stylized assumptions (A) and (B) and addresses empirical omission (C) in prior literature. Results show that an explicit analysis of the time compression tradeoff helps explain classical strategy puzzles including why some innovators freely reveal their intellectual property, how industry leaders systematically fail to stay at the top of their industries, and when early market entry leads to firstmover advantages. These are the broad themes in my work; individual papers differ in their underlying assumptions: Relaxing Assumption (A): Investment Lags in Theoretical Models ◊ Are industries where firms invest faster more competitive? Paper [1] shows that this is not necessarily the case. Competition may be fiercer in industries in which firms are slower to develop new technologies, products, or productive facilities – that is, when investment lags or time-to-build are longer. The lag between investment and production induces more firms to invest under uncertainty. If firms are unable to quickly adjust their strategies to new market and competitive information, the opportunity cost of waiting to invest is higher. Longer time-to-market increases the penalty to firms that do not invest early when demand materializes ex post. Therefore, more firms are willing to make strategic commitments. This result explains anecdotal empirical evidence of chronic excess capacity in industries such as commercial real estate, electricity generation, and bulk chemicals, where it takes a long time to bring new © 2017 G. Pacheco de Almeida

STATEMENT OF RESEARCH ACTIVITIES investments on line. While paper [1] relaxes the assumption of instantaneous investments made in prior literature, it assumes that investment lags are the same for all firms. This is not the case in the papers that follow. Relaxing Assumption (B): Endogenizing Time Compression in Theoretical Models In papers [3, 4, 6, 8, and 9], competing firms decide on the level of acceleration of their activities. Firms are subject to the time compression tradeoff, as in Figure 1. The micromodel of firm investment is sufficiently general to be the workhorse for a broad typology of acceleration problems. My analysis focuses on the speed of innovation and imitation of critical technologies and resources. Three foundational results cut across all papers. First, there exists a right level of acceleration in firm activities – and it is firm-specific because firm capabilities affect time compression diseconomies. Second, time compression diseconomies are a necessary and sufficient condition for sustainable competitive advantage. If speed is costless, rivals imitate instantaneously (and vice versa).ii Finally, sustainable competitive advantage may lead to inferior financial performance. The costs of accelerating activities to be first to market often outweigh the costs of being an imitator – especially when imitators learn from leaders. This last observation is central to papers [3, 4, and 9], which are described next. ◊ Why do some innovators freely reveal their intellectual property? This empirical puzzle has been a focal point of debate in the R&D literature. Papers [4 and 9] show that innovators may share proprietary technology with rivals for free – even if it does not directly benefit them – to slow down competition. By disclosing intellectual property (IP), innovators indirectly induce rivals to wait and imitate instead of concurrently investing in innovation, which alleviates competitive pressure. Free revealing may be particularly useful when what is at stake is not if – but when – rivals develop similar technologies, which seems to be the case in most industries. Thus, innovators should not always "conceal to reduce diffusion" (the classical strategy view), but often "reveal to induce delayed diffusion". This finding is important because it shows that voluntary spillovers – proprietary technology disclosure, patenting in countries with weak IP protection, or geographical collocation with industry followers – should be an optimal strategy more often than expected. Free revealing is an effective anticompetitive strategy that is seldom subject to antitrust law. Intel is an example of an industry leader that may have failed to anticipate these dynamics. Its decision to stop disclosing its post-Pentium technology to AMD did not reduce competition, as intended: instead of waiting and imitating, AMD successfully started racing with Intel to develop the next generation of microchips. ◊ But can free information from innovators ever be ‘too much of a good thing’ for imitators? In contrast with the classical strategy view, paper [3] shows that imitators may not benefit from large interfirm knowledge spillovers. Specifically, imitators may want to limit the know-how that they can freely appropriate from an innovator. Otherwise, innovators have fewer incentives to quickly develop new technologies, which, eventually, decreases the pace and profits of imitation. In other words, if imitators overly reduce innovators’ incentives by appropriating too much of their IP, they essentially ‘bite the hand that feeds them’. Ultimately, this works against them. Thus, imitators may sometimes be better off with lower levels of absorptive capacity. Papers [3, 4 and 9] build on the accepted tenet that rapid imitation reduces leaders’ incentives to innovate. But why does prior work on hypercompetition and Schumpeterian competition argue exactly the opposite: that leader firms should innovate faster when facing rapid imitation? This is the theme of my next paper. ◊ Should leader firms build new advantages faster to retain leadership in hypercompetitive industries? Papers [6 and 8] examine how leader firms should respond to the erosion of competitive advantages caused by rapid imitation and innovation in hypercompetitive environments. On the one hand, shorterlived advantages induce leaders to develop new advantages faster. On the other hand, hypercompetition also erodes the expected returns from new advantages – reducing leaders’ incentives to accelerate investments. Since investing faster also raises costs, this article shows that leaders often prefer to renew competitive advantages more slowly in more hypercompetitive industries – thereby increasing the probability of being displaced by competitors. This phenomenon is dubbed self-displacement. Firms’ decision to self-displace themselves from industry leadership with greater probability is deliberate and rational. The concept of self-displacement is a new explanation for why industry leaders fail to stay at the © 2017 G. Pacheco de Almeida

2

STATEMENT OF RESEARCH ACTIVITIES top of their industries. It is distinct from prior theories of leadership displacement such as organizational inertia, punctuated equilibria, and disruptive innovation, where leaders are said to be displaced because they cannot respond to competitive threats or are not aware of these threats. In this article, industry leaders are aware of and have the means to respond to competitive threats, but simply have fewer economic incentives to remain industry leaders. In other words, investing in sustained competitive advantage is sometimes not profit maximizing. This finding also suggests that firms remain long-term leaders of hypercompetitive industries only by destroying shareholder value. Thus, regular rotation in industry leadership may be a mutually beneficial arrangement to both follower and leader firms in hypercompetitive environments. Papers [3, 4, 6, 8, and 9] focus on acceleration. Working paper [15] contrasts acceleration and experimentation as alternative strategies to quickly respond to new market and competitive information. ◊ When should firms invest in acceleration versus experimentation under uncertainty? Some firms wait until uncertainty is resolved to accelerate investments. Others pursue experimentation ex ante by hedging their investments with a series of small bets in alternative markets or technologies. Both strategies allow firms to quickly respond to market information, but at a cost premium: acceleration leads to time compression diseconomies, experimentation requires costly investments in real options. Thus, are acceleration and experimentation substitute strategies? Only when uncertainty is moderate and investment lags are long, paper [15] shows. With low levels of market uncertainty, firms should not incur the extra costs of either acceleration or experimentation to avoid losses in unfavorable market conditions. However, if uncertainty is sufficiently high, ex post acceleration dominates experimentation: ex ante investments become too risky, even when hedged. Paper [15] is work-in-progress. My next paper [16] also examines experimentation under uncertainty. As in papers [3, 4, 6, 8, 9 and 15], time is endogenous to the model. However, unlike prior papers, paper [16] investigates the microfoundations of strategy dynamics – how it hinges on the assumptions made about individual decisionmaking over time. Specifically, paper [16] uses simulations to study the behavioral dynamics of organizational search in the presence of intuitive biases: ◊ How do intuitive biases affect the process of organizational experimentation and adaptation over time? Paper [16] analyzes when and how organizations can intervene to manage biases and create value in experimentation processes – for example, during an innovation or R&D project. Paper [16] builds on two well-established literatures: it combines the NK model apparatus from organizational search with prior work on intuitive biases in the psychology of human decision making. Drawing on the latter research stream, paper [16] construes biases as unjustified preferences that arise due to automatic, spontaneous thinking. This property of decision making gives rise to a mechanism called generative recurrence. Present this mechanism, unjustified preferences produce two opposing effects on organizational adaptation: they curb excessive experimentation but at the expense of knowledge accumulation. In the context of organizational search, these regularities allow behavioral treatments to strategically leverage the value of biases. Specifically, these results suggest that re-biasing (adopting the opposite bias) often dominates both de-biasing (eliminating the bias) as well as consistently unbiased search. Paper [16] provides evidence that managing rather than eliminating biases can be an effective instrument of behavioral strategy. My last project [19] on auditing verbal theories is a distant spinoff of this research stream. It discusses a procedural methodology and epistemological paradigm to develop and test the robustness of verbal theories in strategy – an analog to the scientific protocol that governs research design in formal economic theory and natural sciences. Since this is preliminary work, no further details are included in this report. Addressing Omission (C): Empirical Analyses of Investment Lags and Time Compression Time remains an unobservable construct in much of the prior literature. The goal of my empirical work has been to measure the time-consuming nature of firm strategy and its effects on strategy fundamentals. ◊ How do investment lags influence the classic strategy tradeoff between commitment and flexibility? Papers [2 and 5] empirically test the theoretical propositions of paper [1]. Using 1975-1995 data from the petrochemical industry in the US, Europe and Japan, papers [2 and 5] study the effect of new plants’ time-to-build on firms’ decisions to invest under uncertainty. Results show that firms are significantly more likely to invest in new facilities when industry average time-to-build is longer. The intuition is the © 2017 G. Pacheco de Almeida

3

STATEMENT OF RESEARCH ACTIVITIES same as in paper [1]. Investment lags slow down firm market entry. Thus, if firms decide not to invest ex ante and there is demand ex post, they will have relinquished profits for the period in which they were out of the market. In this situation, waiting or postponing investments in order to remain flexible becomes a less valuable option. In sum, investment lags offset uncertainty, leading more firms to invest earlier in new market opportunities. These findings support the counterintuitive prediction in paper [1] that industries where firms invest more slowly may actually be more competitive. Papers [2 and 5] focus on the average investment speed in an industry. But when should firms invest faster than the industry? My next paper offers an answer to that question. ◊ Should firms invest faster or slower than competitors? Slow investments cause substantial revenue losses, yet acceleration increases costs. The right speed of investment – and its value – are firm-specific because firm capabilities affect the speed tradeoff. Paper [11] hypothesizes that it is a firm’s intrinsic speed capability, rather than its speed relative to industry competitors per se, that ultimately boosts firm value. This hypothesis finds empirical support in a reduced-form, random-coefficient model of investment acceleration in oil and gas facilities worldwide from 1996 to 2005. Paper [11] shows that the value of speed varies widely among firms and is larger with good corporate governance and less debt. The firm-specific coefficients in the random coefficient regressions show that acceleration rents can be as much as two-and-a-half times more valuable for the best time-based competitors in the sample than for the average of the industry. These results characterize acceleration as competitive advantage’s ‘front-end’ rent-appropriation mechanism. The findings complement previous research estimates of the market value of traditional mechanisms such as patents and trade secrets that appropriate the ‘back-end’ rents of competitive advantage. By empirically elucidating the notion of optimal speed for varied projects, paper [11] also builds a proxy for intrinsic speed capabilities – and links it to firm value. Finally, paper [11] shows that acceleration rents are robust to investment outsourcing. Papers [7 and 10] below build on paper [11] by testing if speed capabilities affect firm entry timing into new markets – and whether this may explain prior mixed empirical evidence of first-mover advantages. ◊ Do intrinsically faster firms wait longer to enter new markets? Entry timing benefits and costs typically vary with firms’ speed capabilities, or the ability to execute investment projects faster than competitors. The main hypothesis of papers [7 and 10] is that firms with intrinsic speed capabilities face low preemption risks and, thus, can afford to wait longer for uncertainty resolution before deciding to enter new markets. This hypothesis is more applicable when investment is associated with higher levels of commitment and, thus, greater option value of waiting. A direct implication is that late entrants with intrinsic speed capabilities should have greater expected post-entry performance. Papers [7 and 10] find support for these hypotheses in the Atlantic Basin liquefied natural gas industry from 1996 to 2007. Empirical paper [14] and project [21] are work-in-progress and address a research question that is central to the tradeoff that initially motivated research streams (1) and (2) on time compression: ◊ How much does it cost to be fast? Prior empirical estimates of time compression diseconomies are sparse and dated. Paper [14] and project [21] aim at contributing to this gap in the literature by measuring the time-cost elasticities associated with time compression diseconomies – or how much costs increase with acceleration – in different industries and empirical settings. Paper [14] empirically investigates acceleration costs in the development of oil and gas production facilities worldwide between 1997 and 2010. The main results of the paper lie in stark contrast with previous studies. On average, the cost of speeding up investments is negative – and not positive, as initially expected: firms could have cut $7.2 million in costs by developing projects one month faster. Over 36% of project time and costs in the sample correspond to unnecessary investment delays and overspending. These estimates are generally consistent with anecdotal evidence from oil and gas industry reports. Importantly, these significant time inefficiencies do not negate the existence of time compression diseconomies per se – rather, paper [14] suggests that time compression diseconomies are often not binding for the vast majority of projects in the sample (87%). This empirical regularity should not be observed in the long-term in highly competitive commodity industries such as oil and gas. Firms with negative time-cost elasticities should either reoptimize operations or exit the industry. Markets with positive time-cost elasticities identify situations with more efficient time-based competition among firms. Paper [14] proceeds to analyze the determinants of time-cost elasticities. These results help reassess firms’ dynamic allocation of capital and the role of time in Strategy theories. Project [21] is a natural extension of paper [14]: it investigates the © 2017 G. Pacheco de Almeida

4

STATEMENT OF RESEARCH ACTIVITIES cost of accelerating technology transfer in the context of the global semiconductor industry. Both paper [14] and project [21] use random-coefficient models to estimate time-cost elasticities. Project [20] is also preliminary empirical work that explores the hypothesized causal link between time compression diseconomies and the sustainability of firm profits. Because of its early stage, no additional information on project [20] is included in this report. Finally, paper [13] is a spinoff of this line of empirical research. It is a technical paper that does not focus on investment lags or time compression. Instead, paper [13] builds on papers [11 and 14] and project [21] above by discussing the novel econometric method used in those studies. Papers [11 and 14] and project [21] are among the first applications of random coefficient models to mainstream strategy problems. Specifically, paper [13] tackles the question: ◊ Can random coefficient models close the gap between theoretical and empirical research in strategy? Empirical work has not caught up with the growing sophistication of strategy theories. Strategy aims at understanding the differential effects of firms’ actions on performance. However, standard regression models only estimate the average effects of these actions across firms. Paper [13] discusses how randomcoefficient models (RCMs) may generate new insights about firm heterogeneity and its effects on performance in empirical settings in Strategy. RCMs allow testing for firm heterogeneity in marginal effects, thereby distinguishing between variables that are statistically significant versus strategically significant. RCMs may also predict firm-specific model coefficients post-estimation. Paper [13] develops a simulation testbed using synthetic datasets to compare the precision of statistical inferences from RCMs versus standard regression models in the presence of firm heterogeneity in marginal effects. Results show that RCMs generally outperform most conventional methodologies. Paper [13] also discusses RCMs’ possible limitations due to sample size requirements and potentially restrictive assumptions. Overall, RCMs allow Strategy researchers to test and build new theories at a more granular level that will further our collective understanding of the basic principles governing competitive advantage. RESEARCH STREAM (3): PUNCTUALITY Modern societies are based on the division of labor. The division of labor is only possible if the actions of economic agents are coordinated in time. Thus, any organizational form trades off the productivity gains from specialization and the costs of time coordination. My research on punctuality – or time coordination – investigates this tradeoff.iii While societal and single-agent punctuality have been studied, organizational punctuality has not. Yet, organizations are the main form of coordination of productive activities. My research examines the reciprocal relationship between punctuality and firm strategy, structure, and performance. Most projects are early stage. Individual papers differ in their level of analysis of organizational punctuality, as follows. Punctuality at the Group Level: The Determinants of Temporal Order My first paper focuses on the simplest type of organizational form: a dyad of socioeconomic agents. This baseline model allows me to study the emergence of temporal order.iv ◊ Are temporal norms only determined by the innateness of culture? Paper [18] focuses on two types of temporal norms, clock time versus event time. Clock time governs western societies and socioeconomic agents typically begin and end tasks at pre-specified times. In event time societies, such as some Latin American countries and parts of the Arab world, agents move to the next task only when the previous one is complete. Contrary to common belief, paper [18] shows that temporal norms may result from economic incentives, not just ingrained cultural practices. The intuition for this finding is best illustrated by a project with two sequential tasks, each performed by a different agent. The downstream agent who performs the second task can follow one of two strategies. First, she can choose to operate in clock time: plan to start the task at the expected time of completion of the first task – and readjust this timing if task one turns out to be unpunctual. Alternatively, she can decide not to schedule the second task in advance, but wait until the first task is completed – that is, work in event time. If the completion time of the first activity has sufficiently high variance (low punctuality), and readjusting the starting time of the second

© 2017 G. Pacheco de Almeida

5

STATEMENT OF RESEARCH ACTIVITIES task is costly, then event time dominates clock time. The opposite holds if the first activity has low variance. Therefore, in paper [18] organizational punctuality is an equilibrium outcome. Punctuality at the Firm Level: Time Coordination and Firm Performance Project [17] is the least advanced in my research portfolio. In its current form, it consists of a preliminary, ‘loosely coupled’ system of ideas on what might affect the specialization-coordination tradeoff. ◊ Does punctuality improve firm performance or can there be a ‘punctuality discount’? One might intuit that the returns to punctuality are always positive. Being on time allows firms to exploit the gains from specialization and to coordinate its exchanges with other firms, which facilitates market transactions. However, there are boundary conditions to this argument. First, punctuality is likely achieved at the expense of speed. To be punctual, firms must build slack time into their operations to accommodate future unexpected shocks, which reduces speed. This curbs the benefits of punctuality. Second, being on clock time may allow less leeway for organizational exploration and innovation, which ultimately reduces firm performance. For example, Google recognized this problem when it launched the ‘Innovation Time Off’ initiative, where engineers spend 20% of their work time on new projects without being subject to hard deadlines. Finally, the returns to punctuality also decrease if other economic agents are unpunctual or market failure limits the division of labor – that is, if time coordination is difficult or unnecessary. Paper [17] aims at addressing these issues empirically. A large dataset on investment punctuality, firm performance, and country-level indicators has been collected. Extensive survey data is also available. Punctuality at the Interorganizational Level: Firm Matching and Relational Rents Firms’ time coordination efforts are strategic complements: the benefits to one firm of being on time increase in other firms’ punctuality. The analysis in my next paper [12] applies to any type of interorganizational complementarity problems – punctuality being just one of many possible examples. Results are reported below in their most general form. ◊ Are synergies from inter-firm complementarities enough to create value in firm partnerships? Yes, according to prior literature. No, according to the simple matching model developed in paper [12]. Interfirm complementarities are necessary but not sufficient for firm partnerships to create value. Some degree of assortative matching – a sorting between the resources of partner companies (i.e., a matching of likes) – is also needed. For example, in time coordination, assortative matching exists when partnerships are formed between firms with comparable punctuality capabilities. It is assortative matching that unleashes the full potential synergies from inter-firm complementarities. This is because the benefits of firm partnerships increase in the quality of the partner (i.e., interorganizational value is supermodular). Thus, relational rents increase in synergies only if there is some degree of inter-firm assortative matching. Paper [12] also shows that, in a frictionless world where information is perfect and managers optimize, firm alliances disproportionately increase the value of high-resource-level firms – resulting in higher variance and higher skewness of the distribution of firm value. Moreover, higher-value alliances are subject to regression to the mean at a faster rate. These effects are magnified if the degree of complementarities is endogenously determined by each firm's investment. Finally, paper [12] uses numerical analyses to show that deviations from the perfectly assortative matching equilibrium – for example, in situations of inter-firm matching with noise or frictions – do not qualitatively change the results. FINAL NOTE ON RESEARCH DESIGN My theoretical papers contribute to the emerging literature on the formal foundations of strategy. The importance of game theory in strategy dynamics has been acknowledged by Management Science (Ghemawat and Cassiman, 53(4), 2002, p. 530): ‘Mathematical models are particularly important in the study of dynamics, because dynamic phenomena are typically characterized by nonlinear feedbacks, often acting with various time lags. Informal verbal models may be adequate for generating predictions in cases where assumed mechanisms act in a linear and additive fashion (as in trend extrapolation), but they © 2017 G. Pacheco de Almeida

6

STATEMENT OF RESEARCH ACTIVITIES can be very misleading when we deal with a system characterized by nonlinearities and lags’. My empirical work shows that basing theories on constructs of inter-subjective reality such as time – which is perceived differently by different people – does not lead to un-testable models and vacuous theory. REFERENCES [1]

Pacheco de Almeida, G., P. Zemsky. 2003. The Effect of Time-to-Build on Strategic Investment under Uncertainty. RAND Journal of Economics 34(1), 167–183. | SSRN

[2]

Pacheco de Almeida, G., J. Henderson, K. Cool. 2003. Resource Accumulation Lags and Strategic Investment Decisions: An Empirical Appraisal. Academy of Management Proceedings.

[3]

Pacheco de Almeida, G., P. Zemsky. 2007. The Timing of Resource Development and Sustainable Competitive Advantage. Management Science 53(4), 651–666. | SSRN

[4]

Pacheco de Almeida, G., P. Zemsky. 2008. Time-Consuming Technology Development: How Imitation and Spillovers Affect Competitive Dynamics. Academy of Management Proceedings.

[5]

Pacheco de Almeida, G., J. Henderson, K. Cool. 2008. Resolving the Commitment vs. Flexibility TradeOff: The Role of Resource Accumulation Lags. Academy of Mgmt. Journal 51(3), 517–536. | SSRN

◊ Reprinted by Maritan, C., M. Peteraf (Editors). 2011. Competitive Strategy. Edward Elgar Publishing. [6]

Pacheco de Almeida, G. 2009. Time Compression in Turbulent Environments. Academy of Management Proceedings.

[7]

Hawk, A, G. Pacheco de Almeida, B. Yeung. 2009. Firm Heterogeneity in Speed Capabilities and Entry into the Emerging Subfield of Atlantic Basin LNG. Academy of Management Proceedings.

[8]

Pacheco de Almeida, G. 2010. Erosion, Time Compression, and Self-Displacement of Leaders in Hypercompetitive Environments. Strategic Management Journal 31(13), 1498–1526. | SSRN

[9]

Pacheco de Almeida, G., P. Zemsky. 2012. Some Like It Free: Innovators’ Strategic Use of Disclosure to Slow Down Competition. Strategic Management Journal 33(7), 773–793. (Lead Article) | SSRN

[10] Hawk, A., G. Pacheco de Almeida, B. Yeung. 2013. Fast-Mover Advantages: Speed Capabilities and Entry into the Emerging Submarket of Atlantic Basin LNG. Strategic Management Journal 34(13), 1531–1550. | SSRN [11] Pacheco de Almeida, G., A. Hawk, B. Yeung. 2013. The Right Speed and Its Value. Strategic Management Journal 36(2), 159-176. (Lead Article) | SSRN [12] Cabral, L., G. Pacheco de Almeida. 2016. Alliance Formation and Firm Value. Management Science. Conditionally accepted. [13] Alcácer, J., W. Chung, A. Hawk, G. Pacheco de Almeida. 2017. Testing for Firm Heterogeneity and Predicting Firm-Specific Coefficients Using Random Coefficient Models. HEC Paris Working Paper. | SSRN

[14] Hawk, A., G. Pacheco de Almeida. 2017. Time Compression (Dis)Economies: An Empirical Analysis. HEC Paris Working Paper (in preparation for 3rd review round resubmission, Strategic Management Journal). [15] Pacheco de Almeida, G. 2013. Acceleration vs. Experimentation. HEC Paris Working Paper. [16] Korniychuk, A., T. Obloj, G. Pacheco de Almeida, E. Uhlmann. 2017. Re-Biased Search: Managing Intuitive Preferences over Time. HEC Paris Working Paper (in preparation for resubmission). | SSRN [17] Punctuality (with L. Cabral). [18] Clock Time vs. Event Time (with L. Cabral). [19] Auditing Verbal Theories (with T. Obloj). [20] Time Compression and Profit Persistence (with A. Hawk) [21] The Cost of Accelerating Technology Transfer (with A. Hawk)

© 2017 G. Pacheco de Almeida

7

STATEMENT OF RESEARCH ACTIVITIES

FIGURE 1 10

REVENUES AND COSTS

Revenues

5

Profits

Costs 0 0

10

20

30

TIME Optimal Time

The time compression tradeoff: slow investments cause revenue losses, but acceleration increases costs. Firm capabilities and inter-firm knowledge spillovers affect time compression diseconomies; competition affects revenues. i

Time compression diseconomies have two main causes. First, acceleration usually involves crash investments, with the deployment of more resources to a project at each point in time. The law of diminishing returns – where one input, viz. time, is held constant – limits productivity and drives up investment costs. Second, acceleration generally calls for parallel processing of previously sequential activities. This reduces information flows across different stages of the project development process, increasing mistakes, rework, and costs. ii

If investments can be accelerated at no cost, imitators copy instantaneously: time compression diseconomies are a necessary condition for sustainability. Conversely, if there are time compression diseconomies, the cost of instantaneous investment is infinite – it is impossible to develop projects on the spot unless they are traded on open markets. Thus, time compression diseconomies are also a sufficient condition for sustainable competitive advantage. iii

Punctuality bears the hallmarks of coordination games. Coordination requires temporal precision: punctuality is more than just not being late – it is also not being early vis-à-vis preset expectations. My work defines punctuality (from Latin punctum, point) as the quality of an action performed at a pre-specified time. Any deviation from this temporally predicted point entails economic costs. Being early to a scheduled meeting is detrimental because of the opportunity cost of idle time, the diseconomies incurred by performing preceding activities faster than necessary, and the inefficiencies from interruptions in the work flow. On the other hand, the costs of tardiness include the foregone revenues from missed economic transactions, losses in aggregate productivity, and the penalties associated with rescheduling and reorganization. Finally, note also that, in exceptional circumstances, individuals may be unpunctual but still coordinated in time as long as there is no within-group variance in timing (e.g., when everyone is equally late to a meeting). iv Historically, punctuality was unknown to the Ancient and Medieval civilizations, but it was also not an invention of modern society. Punctuality largely preceded the development of accurate mechanical clocks. It started as an individual ethical virtue in the 16th century: this new conception of time was a strictly spiritual approach to God, most valued in the Protestant world of Calvinist faith. It was in Geneva that punctuality first permeated public life through institutionalized protocols of social control (see Max Engammare, “L’Ordre du Temps”, 2004). This fact may explain the early rise to dominance of the clock industry in Switzerland. Although punctuality was invented in the 16th century, it only became a widely accepted temporal norm in secular societies in the late 18 th and 19th centuries with the industrial revolution, time standardization, and the rise of rationalism and Newtonian physics.

© 2017 G. Pacheco de Almeida

8