August 15, 2023

Can economic growth be understood by relying only on the insights of one discipline? Or does our understanding increase markedly if insights from many disciplines are integrated?

Introduction

With billions of people still living in poverty in the world, there is perhaps no more important question in human science than what are the causes of economic growth. Moreover, it is a very complex question, for economic growth is influenced by interactions among a host of economic, political, social, cultural, and geographical phenomena. This chapter discusses how a process for interdisciplinary research can usefully be applied to the study of economic growth. It is thus simultaneously an exploration of how to do interdisciplinary social science and how to develop a more comprehensive understanding of economic growth. It shows how a variety of distinct research programs across all social science disciplines can be integrated to enhance our understanding of the causes of economic growth. This chapter is organized according to the steps in the interdisciplinary process.1 Although the chapter will review each of the various steps ideally involved in interdisciplinary analysis, the presentation focuses primarily on how “common ground” can be achieved among disciplinary insights that conflict, notably with respect to the role of government, the role of international trade relationships, and the process by which economic institutions are and should be developed. Lessons are drawn for each step regarding both our understanding of economic growth and how best to perform interdisciplinary research.

AUTHOR’S NOTE: My chapter summarizes and comments upon my book The Causes of Economic Growth: Interdisciplinary Perspectives (Szostak, Rick, 2009, Berlin: Springer). The question addressed in this chapter is both very complicated and very broad. Both characteristics increase the difficulty of performing any of the steps as completely as one might like (though perhaps especially the literature survey). The question is complicated because a variety of phenomena combine to influence economic growth. By asking the question at the general level rather than with respect to a particular time and place, we broaden the scope of inquiry significantly: We could forget about questions of basic property rights if we focused on only the rich world, for example. That is, the breadth of the question forces us to engage its full complexity. We thus need a broader literature survey, and we need to identify, evaluate, and integrate across a much broader set of disciplinary insights than would normally be the case. One possible exception may be the first step: It may be just as easy to frame a broad question as a narrow one. Indeed, it may be easier, for the narrow question requires detailed definition of the boundaries of the question. Identify an Interdisciplinary Research Question We have selected our question: What are the causes of economic growth? It is then necessary to ask whether

SAGE

SAGE Research Methods

Page 2 of 30 An Interdisciplinary Analysis of the Causes of Economic Growth this question is suitably interdisciplinary in nature. Can economic growth be understood by relying only on the insights of one discipline? Or does our understanding increase markedly if insights from many disciplines are integrated? Economic growth must, in the first instance, involve an increase in the resources devoted to production—broadly, labor, capital, and natural resources (including land itself)—and/or the productivity with which these resources are combined to produce output. These four variables—labor, capital, natural resources, and productivity—are commonly termed the proximate causes of growth, and economists (and economic historians, who are treated as a separate discipline here) dominate the study of the question of which of these is most important in driving particular episodes of economic growth. As we shall see, economists are far from achieving consensus on this basic question. Moreover, economists have long appreciated that this question then invites a more complicated set of questions, such as the following:

• Why is labor more skilled in some countries than in others?

• Why is there more (saving and) investment in some countries than in others?

• Why do some countries use resources more productively than do others?

• Why is productivity higher in some countries than in others?

These questions tend naturally to invite interdisciplinary speculation. How does a society’s culture or social structure or politics influence its educational attainment, work effort, saving rate, or environmental policy? The study of these deeper causal influences is pursued across the human sciences (see below). Special mention should be made here of institutions and technology. The formal rules of a society—its legal system, economic regulations, firm structure, and so on—have a profound influence on its economic performance, and yet such institutions arguably (see below) emerge from a historical process involving political, social, and cultural influences. Likewise, technological innovation is an important source of (at least modern) economic growth, and again, it seems likely that the rate of innovation in a society may well be influenced by a host of non-economic factors.

Even the statistical analyses of economists point toward interdisciplinary analysis. Political, institutional, and social variables are often found to be important in cross-country analyses of postwar growth experience (Snowdon, 2002, pp. 97–99)—and this despite the twin facts that such variables are often hard to measure and likely exert their effects over a very long time. Surveying this evidence and the widely divergent growth experiences of postwar economies more generally, Snowdon concludes:

To understand why some countries have performed so much better than others with respect to growth it is therefore necessary to go beyond the proximate causes of growth and delve into the wider fundamental determinants. This implies that we cannot hope to find the magic bullet by economic analysis alone. (p. 100) The observation that it is crucial to look beyond proximate causes provides an important insight regarding interdisciplinary research more generally. Economists have, until recently, been able to view the causes of economic growth as a strictly disciplinary question by looking only at the interaction of a handful of economic

SAGE

SAGE Research Methods

Page 3 of 30 An Interdisciplinary Analysis of the Causes of Economic Growth variables. Interdisciplinarians need to be sensitive to the precise wording of their focus question (and be prepared to revise it as they perform later steps), in order to ensure that relevant disciplines are not arbitrarily excluded from examination. Even though we are striving to explain movements in an economic variable, our question becomes interdisciplinary once we embrace a wide range of potential causes. Identifying Relevant Phenomena, Theories, Methods, and Disciplines The next step or steps must involve the gathering of relevant disciplinary insights. How does the researcher know where to look? There are two complementary strategies identified in Repko (2008). One is to reflect on the character of different disciplines and identify those that are likely to have something to say about the issue of concern. The second—and the one pursued in this chapter—is to ask what phenomena, theories, or methods are implicated, and then ask which disciplines study each phenomenon identified and/or apply each theory or method. This approach reduces the risk of favoring the larger and most familiar disciplines (Szostak, 2002). Szostak (2004) develops exhaustive classifications of phenomena, theory types, and methods to facilitate the latter approach: In the absence of these, it is all too easy to assume that the subset of relevant theory, method, or phenomena pursued by disciplines is somehow appropriate. Repko (2008) identifies the defining elements of disciplines in terms of these and other classifications. One challenge the interdisciplinary researcher will face is that library catalogues are organized by disciplines, and different terminology is used in different disciplines to refer to the same phenomenon, theory, or method (Szostak [2007, 2008] addresses how a classification suited to interdisciplinarity might be developed). In the case of growth, it is embarrassingly easy to identify phenomena that, at least potentially, influence growth but lie outside the (at least until recently) narrow gaze of economists. These include cultural attitudes, political institutions, geographic constraints, and ethnic tensions, among others. It is thus straightforward to implicate (parts of) all social science disciplines as well as the humanities in this study. As for types of theories, economists rely almost exclusively on methodological individualism: Only individuals are causal agents. Yet, surely the alternatives of relationship or group agency2 matter for at least some of the proximate causes of growth. Historians have long since abandoned the idea of the heroic innovator working in isolation in favor of an appreciation of the networks in which innovators operate; the same logic applies to entrepreneurship and trade more generally, and surely institutional change cannot be fully appreciated without recourse to relationships and groups. Economists also stress rational decision making. Yet, the history of institutional change suggests that rationality likely plays some role in institutional design—agents consciously design institutional improvements—but also that tradition does—agents are cautious in moving away from existing institutions. Historians of both science and technology have long appreciated that a mix of reason and intuition is involved: The investigator consciously gathers relevant information, but insight comes subconsciously as novel connections are drawn. Moreover, in a world awash in information, who can doubt that economic agents often follow decision-making rules (“Buy when the market is rising”) rather than attempt rational

SAGE

SAGE Research Methods

Page 4 of 30 An Interdisciplinary Analysis of the Causes of Economic Growth calculations? Perhaps most important, both investment and innovation decisions must be made under uncertainty—people simply cannot know the likelihood of particular outcomes—and people cannot fully rely on rational decision making when faced with uncertainty (as economic theory admits), but necessarily follow hunches or decision rules or mimic others. Various sorts of nonrational decision making have long been studied outside economics, especially in sociology. Psychologists have long argued for different types of decision making; brain imaging shows that different parts of the brain are activated at different times and make decisions in different ways (Cohen, 2005). Economists model growth solely in terms of steady-state (constant) growth rates. Yet, growth occurred in the West much more rapidly in the 19th century than ever before and more rapidly in the first postwar decades than before or since. In both the 19th and 20th centuries, one can discern multiple periods of a decade or so in length in which growth was relatively slow (by the standards of those centuries) or negative. The growth experience of other regions of the world is even more diverse. These diverse experiences suggest that theories with either cyclical or stochastic elements—allowing growth rates to both rise and fall—should be important. The fact that growth rates are, at least potentially, much higher than a mere two centuries ago (or alternatively, common prognostications that growth will soon decline) suggests that theories positing dynamic change in one direction may also have a role to play. Various theories more commonly employed outside rather than within economics are thus important in understanding economic growth (though each of these has its own limitations). Alternative theories worth exploring include the following:

• Evolutionary theory can potentially embrace all types of agency, decision making, and time path.

• Systems theories (or structuralism/functionalism more generally) can also reflect a variety of types of agency and decision making. In practice, systems theorists have tended to emphasize system stability (equilibrium) and thus have had less to say than they might about dynamic processes of change.

• Social constructionist theories stress the importance of attitudes and beliefs. This provides a useful complement to the focus of most social science theory on actions, though social constructivists often see their theories as substitute rather than complement.

• Modernization theories in the early postwar period posited that all countries would move toward “Western” economic, political, and cultural realizations. These gave way to more pessimistic dependency and world systems theories (inspired in part by Marxian analysis) that suggested that poor countries would remain poor. These theories can each be valued for detailing possible links among economic, political, cultural, and other phenomena without adopting their assumptions about the inevitable outcome of these systems of interaction.

• Complexity theories can generate all types of time path. They are perhaps particularly valuable in stressing stochastic outcomes. Complexity theories often emphasize system-level emergent properties at the expense of careful specification of individual causal links.

• Various theories stress how culture shapes behavior. These theories are often characterized by

SAGE

SAGE Research Methods

Page 5 of 30 An Interdisciplinary Analysis of the Causes of Economic Growth vague terminology and fail to appreciate how cultures evolve.

• Psychological theories support the insight that decision making is not always rational, but these theories have not often been carefully applied to economic decisions. (The emerging field of behavioral economics aspires to change this.)

• Literary theory may also be useful. In studying technological diffusion, scholars have long appreciated that people, not just blueprints, generally have to move from one locale to another. There is a range of tacit knowledge that is imperfectly captured by the most careful instructions. This observation is the same as that long made by theories of texts in general: There is always a divergence between the textual signifier and that which it is presumed to signify. A similar analysis could be performed with respect to research method. One of the key insights of Szostak (2004) was that each method is better at investigating some theory types than others, and disciplines thus choose mutually compatible sets of theory and method (and phenomena). Various case study methods—observation, textual analysis, interviews—are better suited to the investigation of many of the theories listed above than are the statistical analyses favored by economists. Moreover, different methods shed different light on different theories: We should have the greatest confidence in a theory that is supported by different methods. When examining a complex historical process such as economic growth, which involves many causal interactions, recourse to multiple methods is particularly important. There are four criteria for identifying a causal relationship: establishing correlation, establishing temporality (the cause should generally appear before the effect), ruling out alternative explanations of the result, and showing how the causal relationship unfolds in practice (including identifying intermediate variables; Singleton & Strait, 1999). The methods of economists excel with respect to the first two. Their lack of attention to alternative theories limits the third. With respect to the fourth, economists are often attracted to mathematical models even when the evidence for particular causal relationships is quite limited. The latest type of economic growth models—called unified growth models—attempts to model the course of economic performance over the last millennium. These models posit that small changes (in the first models, changes in population density, but later models treat other variables) eventually surpass some threshold where they begin to have dynamic effects on growth. But of course, one can develop mathematical models such that any cause can have any effect, if one assumes that small changes achieve big results. The creation of these models should not enhance one’s confidence in such a relationship in the absence of careful case study evidence. Disciplinary perspectives will be treated briefly here. As both Szostak (2003) and Repko (2008) have stressed, preferences with respect to theory and method and phenomena are critical components of disciplinary perspective. Ideological, ethical, and epistemological predispositions need also to be appreciated in evaluating disciplinary insights. In these latter respects, the following can be noted:

• On average, economists believe in markets more than other social scientists do. Although economic theory suggests a variety of market imperfections that may arise, the average economist may, nevertheless, tend to downplay the role of governments in the process of economic growth. In turn, other disciplines may underestimate the role of markets in fostering growth.

SAG

SAGE Research Methods

Page 6 of 30 An Interdisciplinary Analysis of the Causes of Economic Growth

• Economists are generally consequentialist in ethical orientation, while other disciplines place greater emphasis on tradition, virtue, or intuition. Economists in particular tend to think that economic growth is good, while scholars in other disciplines are more likely to critique at least some elements of growth. Although questions regarding the desirability of growth can be distinguished from questions about causes, thoughts about one naturally influence thoughts about the other.

• Economists tend to be realists and assume that scholars can obtain reasonably accurate understandings of a fixed external reality. Other disciplines, especially in the humanities, cast a useful, if often exaggerated, skepticism on the possibility of human understanding. They thus encourage scholars to be more careful in both their theorizing and policy advice. In particular, these other disciplines are suspicious of broad generalizations and encourage careful context-dependent research. The preceding analysis has illustrated the following points regarding these steps in interdisciplinary analysis:

• Interdisciplinarians should be careful of curtailing the scope of their research unwittingly by following major currents in the existing literature. In the case of economic growth, which other disciplines have rarely stressed as a topic, it is particularly likely that scholars may produce valuable insights that are relatively unheralded within their own discipline.

• Nevertheless, interdisciplinarians can usefully focus on particular disciplinary subfields. The danger of missing relevant literature will be reduced if they also reflect on what theories and methods might be relevant to the issue at hand.

• Interdisciplinarians must evaluate disciplinary insights in the context of disciplinary perspective and with attention to the (complementary) strengths and weaknesses of different theory types and methods.

Evaluating Disciplinary Insights

The interdisciplinary researcher must next evaluate the disciplinary insights generated by the relevant theories and methods. This is an exercise in critical thinking. Interdisciplinarians must know how to distinguish argument from assertion and assumption from evidence. In addition to standard strategies for the critical analysis of any text, interdisciplinary analysis suggests several important strategies for critique:

• In what ways might a particular disciplinary insight be shaped by the particular “perspective” of that discipline?

• More concretely, how might the insight be altered if the researcher(s) had examined a wider set of phenomena?

• Likewise, what are the strengths and weaknesses of the theories and methods used by the discipline, and how might the insight in question be shaped by these?

• Do the insights of one discipline point to possible weaknesses in the insights of another? Insights from outside the academy can also be quite useful here.

SAGE

SAGE Research Methods

Page 7 of 30 An Interdisciplinary Analysis of the Causes of Economic Growth

The analysis of the causes of economic growth provides examples of each of these evaluative strategies. As we have seen, economists are likely to stress the role of individuals and rationality and may thus overlook the actions of groups and/or various types of nonrational behavior. Economists have focused on a narrow set of economic variables while downplaying the importance of, for example, culture. Economists stress equilibrium or steady-state outcomes in their modeling exercises (in large part because this makes the math more tractable, but also because of the emphasis of economists’ general equilibrium theory on system stability); in the real world, of course, growth has never been steady. Economic historians, sociologists, and political scientists often stress that different countries have differing experiences of economic growth; this provides a useful counterpoint to the economist tendency to identify central tendencies. The interdisciplinarian can be heartened by the observation that other disciplines avoid at least some of these biases in economic analysis. Yet, the interdisciplinarian should never forget that all disciplines have limiting perspectives:

• The other social sciences have long stressed group or relationship agency (or viewed individuals as constrained to act in a certain way by culture or institutions), without detailing how these constraints emerge. In sociology, this approach has, in recent decades, been supplemented by individual-level analysis, but syntheses of these approaches are rare. The interdisciplinarian must be prepared to integrate insights that have rarely been juxtaposed in the past.

• Various sorts of nonrational behavior have been investigated in sociology and other social sciences. As in economics, particular types of decision making are often assumed rather than established empirically. The interdisciplinarian must thus be prepared to reflect on what sorts of decision making motivate people in particular situations. This task is made more difficult by the fact that these are often mixed in practice: An investor may only act if gut instinct, rational calculation, and the actions of others all point in the same direction. The analyses of many social scientists present a further challenge: They celebrate the “irrationality” of certain behaviors without carefully identifying what sort of nonrational decision making is at work.

• The social sciences (including psychology) collectively employ each of the dozen methods utilized in the scholarly enterprise. Yet, each of these is applied to only a subset of appropriate questions. The interdisciplinarian, aware of the biases of one method, may look in vain for the application of alternative methods to particular questions. For the present inquiry, this problem—of difficulty in identifying work that applies different methods to the same question—is exacerbated by the fact that scholars in other disciplines have rarely addressed economic growth itself.

• Other social sciences often assume processes involving sustained change in particular directions, just as economists assume equilibrium outcomes. The important possibility of stochastic (unpredictable) outcomes is much less commonly explored. The growth process likely entails elements of each. Investment decisions at times seem to follow herd behavior and at times seem inexplicable, and yet there is an amazing stability (lack of volatility) in average rates of return over time. As with types of decision making, the interdisciplinarian generally has recourse to few previous attempts to integrate insights reflecting these different perspectives.

• Political scientists and sociologists often assume the superiority of governments over markets. Only a minority of scholars in any discipline carefully compares the advantages of one or the other for particular types of decisions (e.g., identifying the reasons why science is publicly funded but technology is largely left to markets and the potential difficulties with each approach), and only a minority appreciates that the ideal balance likely varies across time and place. The lesson for interdisciplinary practice here is that the evaluative step should not be conflated with the next step of finding common ground (even if, in presentational terms, it proves useful in this chapter to discuss evaluation while outlining common ground). That is, disciplinary insights should not just be critiqued when they conflict. The mere fact that different disciplines have asked different questions, and thus often not directly disagreed with each other, should not render the interdisciplinarian sanguine about these disciplinary insights. Interdisciplinarians can thus suggest useful clarifications or extensions to theories even in the absence of a direct contrast among disciplinary insights. Creating Common Ground The almost-final step(s) involve finding some “common ground” that integrates (elements of) various disciplinary insights. The role of creativity, intuition, and inspiration may loom large here. However, certain straightforward techniques can be applied to find common ground:

• One can first ask to what extent seeming differences in disciplinary perspective are apparent rather than real: Differences in terminology may mean that different disciplines are not actually talking about the same causal process even when they appear to be. The interdisciplinarian can often redefine concepts or extend a concept from one discipline to the subject matter of another.

• When concepts conflict, they can often be placed on a continuum: The tendency of economists to stress rationality and of sociologists to stress irrationality can be handled by evaluating the degree of rationality that individuals may display in a certain situation (Newell, 2007).

• One can then ask whether remaining differences can be overcome by small alterations in disciplinary assumptions.

• The easiest path to creating common ground involves relaxing the assumption made by each discipline that only its phenomena matter: Economists try to explain growth by focusing on a handful of economic variables, while sociologists stress matters of culture and social structure (happily, these attitudes are starting to change). In this simplest situation, the insights of different disciplines can often be added together: Innovators may respond both to the economic incentives stressed by economists and the cultural values emphasized by sociologists. Different disciplines are highlighting different aspects of the question under investigation, but they are assuming other aspects do not exist. Of course, integration is still required in order to identify how these different causes interact in different circumstances.

• When different disciplines reach different conclusions regarding the same phenomena, the problem is often one of excessive generalization, and it can thus be solved by more carefully expressing the range of applicability of the theories involved. Examples of each technique can be readily provided. Economists use the word “investment” quite differently from business scholars, for example: To economists, it means the actual production of buildings or equipment that are then used in the production of other goods or services. Business scholars are likely to include any effort to make a profit, such as speculation in various markets, in the term investment. Economists have ignored certain things such as the influence of culture, the importance of networks, and the uneven path of technological innovation. At times, the insights of other disciplines in these areas can simply be added to the insights of economists—after each has been carefully evaluated in turn. Economists often assume that a particular institution is optimal, while sociologists and political scientists show that institutions operate in a complex web of social interactions: The different traditions can together strive to identify how well different institutions serve economic growth in different contexts. Rather than organizing this section in terms of the type of integration pursued, it will be more useful to organize it by topic. Common ground will be found along a variety of different causal links generating economic growth. Within each topic, though, care should and will be taken to identify the integrative strategies utilized. This approach—examining different causal links in turn (but not losing sight of emergent properties of systems)—is different enough from the standard practice in economics of formally modeling several links at a time that it deserves some comment. There are difficulties in applying the methodology of models to open systems—ones which are clearly linked causally (as the economy surely is) to phenomena outside the model. Although these problems do not destroy the model-building exercise, it is nevertheless true that such models rely on an unrealistic assumption that the relationships observed in a system can remain fixed through time. Notably, complexity theory takes a different approach, allowing different causal forces to operate along different links, and does not assume any particular organizing principle (such as equilibrium) at the outset. It should also be noted that this emphasis on causal links is not common to all efforts to identify best-practice processes of interdisciplinary analysis. Interdisciplinarians often stress the importance of emergent properties (e.g., Bammer, 2005). It is argued here that the two approaches are complementary and thus best pursued in tandem. This strategy accords well with the general inclination toward integration by interdisciplinarians. As well, interdisciplinarians have often said vaguely that there are different “facets” to complex problems. The emphasis here on causal links provides a means of clarifying what might be meant by the vague term “facet.” Since different causal links tend often to be the focus of different disciplines, the strategy of placing diverse causal links within an overarching structure can be a powerful technique for interdisciplinary analysis. The causal link approach helps us identify when different disciplines are, in fact, speaking about the same thing, and thus it sets the stage for integrating disciplinary insights link by link.3 The Proximate Causes Themselves Economists have devoted an enormous amount of effort to “growth accounting” over the past decades. These empirical exercises attempt to identify the relative importance of the proximate causes of growth: How much is due to investment as opposed to innovation, for example? These exercises have been valuable: In particular, economists in the 1960s were shocked by the fact that investment in physical capital accounted for only about a third of economic growth over the previous century—and they were guided to pay more attention to education and innovation as a result. Yet, these exercises rely on a rarely voiced assumption: The effect on growth found for each proximate cause in one study should at least be a central tendency for all economies at all times. This assumption is dangerous: Easterly (2002), an applied economist, describes how the World Bank was led to a number of naïve policies over the years as a result, such as calculating the “required investment” needed for certain target rates of growth and channeling those sums into countries ill-prepared to utilize them productively. Historians and economic historians have stressed the particularities of different cases. Economic historians have long hypothesized that different generations of industrializers faced different challenges, and thus that they necessarily developed in different ways (The classic argument was that of Gerschenkron [1962]; see Sylla & Toniolo [1991] for an update). The same concern has been voiced, albeit using quite different theories, by dependency and world systems theorists in sociology and other social sciences. But these scholars did not (to my knowledge) directly address the assumptions of growth accounting, and thus it remains for the interdisciplinarian to make the connection. The assumption has a corollary: The proximate causes act independently. The main reason that the World Bank strategy failed was that the return on investment in a country depends on many things: levels of education, infrastructure, technology, and so on. Complicated relationships among such supposedly independent variables are hard to capture within standard statistical procedures. That is, it is easy to estimate how investment affects growth and education affects growth, but it is harder to establish how they combine to do so. And thus a more nuanced understanding of how proximate causes interact will depend on detailed case studies.

This adjustment in our understanding of the relationship between investment and growth is entirely in accord with the strategies for interdisciplinary analysis outlined above: Growth accounting regressions naturally omit many variables that condition this relationship; the structure of those regressions does not allow for independent variables to act in concert; the approach reflects a disciplinary tendency to identify supposedly enduring causal relationships without careful concern for the set of conditions in which these might hold; and the growth accounting analyses thus represent a widespread tendency in scholarship to assume greater generality for one’s results than they deserve.

Trade and Growth

Solow (2005, p. 4) argues that there has been less modeling of growth in open economies than he would have expected 50 years ago. Growth models tend to focus entirely on the internal dynamics of the growth process. This is unfortunate. Static economic analysis suggests that free trade is good because by pursuing their comparative advantage, countries will be able to increase the value of their total output. They will export goods that they are relatively good at producing and import goods that they are relatively poor at producing. This is one of the most powerful insights in all of economics. But what if productivity (output per labor and capital input) is advancing fastest in the goods a country imports? It will still benefit by being able to import more as the international price of these goods falls, but it will not benefit as much as it would if its own economy was experiencing rapid productivity growth. And thus there is a theoretical possibility that a country may benefit in the long run by sacrificing some of the short-term gains from importing and exporting in order to encourage production of goods where productivity advance is more likely. Empirical research by economists has often suggested that the benefits from openness to trade are much greater than static theory can explain: They thus point to dynamic benefits such as increased technology transmission or greater competitive pressures on local firms. Yet, these empirical results have been questioned even by economists, who note that they are not always obtained when different definitions of “openness” are used or different time periods investigated. Economic historians, sociologists, and political scientists have stressed that all successful developed economies have been protectionist early in their development process. They have theorized that protectionism encouraged growth (see Miller, 2008). Most economists will not readily accept a suggestion that openness to trade is not necessarily a good thing (even this author finds it difficult). Yet, we have just seen that economists have proven willing to accept arguments that the dynamic gains swamp the static comparative advantage gains. If arguments for huge dynamic advantages are allowed, then arguments for dynamic disadvantages cannot simply be ignored. Given the limited degree of theorization of dynamic effects, heavy weight must be placed on the empirical evidence. The statistical analysis is not conclusive. The historical evidence points toward a role for protectionism. But the historical record is also full of failed efforts to protect, of governments that allowed protected industries to focus their energies on maintaining government support rather than becoming internationally competitive. Case study analysis is tricky here: It is much easier to identify the many failed infant industries—those that received protection from government but failed to grow up—than to observe industries that developed behind tariff protection and establish that they could only have done so behind such protection. In the second case, a counterfactual—what would have happened without protection—needs to be carefully tested. Although both theoretical and empirical analyses are thus more muddled than we might like on the grand question of openness, some answers to narrower questions seem clearer:

• If a country protects, it must ensure that firms face clear incentives to enhance productivity.

• Countries with incompetent and/or corrupt bureaucracies should be particularly wary of managed trade.4

• The inflow of information about technology and institutions should be encouraged, and trade in goods is one way of doing so.

• Tariffs are not the only, and perhaps not the best, strategy for supporting industrial development. These important insights (which emerge from integrating insights from different disciplines) can be thought of as “common ground” in the debate between advocates of openness and advocates of trade barriers. That is, although it is difficult at present to sketch a common ground on the larger question of whether free trade is good for growth, it is at least possible to identify certain circumstances under which this result is more or less likely. These results, although limited, nevertheless carry important policy implications.

Technology and Growth

As noted above, economists were surprised when early efforts at growth accounting showed that investment in physical capital could account for at most a third of the growth observed in developed countries. Economists were guided to look elsewhere and came to stress the importance of education (i.e., investment in human capital), technology, trade, and institutional changes that might enhance economic efficiency. The earliest growth models treated technological innovation as exogenous: something that happened outside the models. The latest generation of endogenous growth models try to bring technological innovation inside the models; they argue that technological innovation results from another sort of investment—in research and development—and thus can be explained in terms of economic variables. These models have been valuable in encouraging economists to devote greater attention to the causes and effects of technological innovation. Yet, the tendency to assume that the rate (and direction) of innovation can be understood entirely as a function of economic variables carries the obvious danger that other influences on innovative activity will be ignored. Scholars of technology, whether historians of technology or scholars of science and technology studies (STS), have certainly stressed a wide range of noneconomic influences on both the rate and direction of innovation. Most obviously, historians of technology have traced how each innovation builds on preceding innovations. Opportunities for innovation at any point in time (and space) are thus conditioned by the existing body of knowledge. This insight, long appreciated in economic history, has begun to be voiced in economics itself. The literature on General Purpose Technologies stresses that occasional “big” innovations set the stage for a series of minor but cumulatively important follow-up innovations. There was thus more innovative potential in the decades after the development of internal combustion (or steam) engines than in the decades before. Not only was there scope for many small improvements to these engines, but they were applied to an increasing range of uses, from factories to cars and planes. Economists have shown less interest so far in the variety of other causal links identified by scholars of technology.5 It is often thought—and not without cause—that STS scholars often assume, rather than identify, cultural influences. When STS scholars claim that innovation is entirely a cultural product, unconstrained by whether it reflects how the world works (an argument more often made with respect to science than technology), skepticism is invited. This should hardly restrain other scholars from respecting more nuanced arguments, and the evidence for them, that culture interacts with technological potential—and a host of other psychological, political, and social influences—to determine especially the direction but also the rate of innovation.

In the case of technology, different disciplinary insights can largely be integrated by dropping the “only these things matter” assumption. Different scholars have examined different causal influences on innovation, all of which likely have a role to play. The popularity of extreme assumptions—that only economic, or cultural, or technological influences matter—serves as a warning that scholars may assume rather than establish the importance of different causal links. And thus, the interdisciplinarian attempting to evaluate the relative importance of different links will need to carefully evaluate the arguments of all scholars. And he or she will also need to be sensitive to the fact that—as with growth above—the relative importance of these links likely changes across time and space. Meeus and Hage (2006) edit an interdisciplinary handbook on innovation precisely because they believe that management specialists, economists, sociologists, historians, political scientists, and others need to integrate their efforts. They note that political scientists tend to stress government policies, management scholars look at firm behavior, economists and sociologists emphasize industry-level analysis, and STS scholars stress scientific innovation (but that until recently, very few scholars in any discipline have looked at behavior in research laboratories); they urge a co-evolutionary perspective in which the interactions among different levels of analysis are appreciated (p. 4). Institutions Do the right sort of institutions encourage growth (and if so how), and which sorts of functions are most necessary for institutions to perform? Because there is abundant evidence that institutions are indeed important, the question then arises how beneficial types of institutional change can be encouraged. How important are institutions? Although economists have, in recent years, included many institutional measures in growth regressions, it has proven difficult to establish any relationship empirically. This outcome reflects, in part, the facts that different economists emphasize different theoretical arguments and that they disagree over the precise definition of institution and which particular institutions might be most important. Measures of social capital and social structure (see below) are often deemed “institutional.” These analyses emphasize political over economic institutions, yet the former, for the most part, influence economic growth only indirectly through the latter. Although there may be direct effects of political stability (e.g., on investor confidence), the primary effect of stability and especially more mundane questions of electoral practices or executive powers will be to shape the nature of economic institutions. Why not, then, stress economic institutions in growth equations (with perhaps other equations linking economic and political institutions)? A further problem is relevant here: It is all too easy for incompetent or corrupt governments to create the appearance, but not the essence, of good institutions. Just as autocrats often glory in meaningless elections, so in the economic sphere they can proclaim the protection of property while actively interfering with property rights. They might expropriate property from citizens through fake legal proceedings or simply fail to enforce private contracts. Economists thus stress the quality of institutions (but a very limited set of them). But, of course, quality is always hard to measure. Finally, the various institutional variables may interact with each other and reflect feedback effects from growth itself, but simple regressions ignore these possible effects (see Aron, 2000). Because different variables are strongly correlated, different economists identify different institutional variables as most important (Lal, 1988). Moreover, Aron (2000, pp. 100–101) worries that statistical analyses of growth in general face serious problems of data, methodology, and identification. These are generally more severe than statistical analysis elsewhere in economics. Although some of the problems identified by Aron can be alleviated by more careful definitions and theoretical specifications, others cannot. Institutions are not particularly well-suited to statistical analysis. Institutions are inherently unique. Countries may differ along various dimensions: The courts may be fair but the police incompetent in one country, while rights may be enforced without much recourse to courts in another. Such complexities call for comparative case studies and suggest that efforts to describe a type of institution in terms of one numerical indicator are likely to be fraught with difficulty. Economic historians have, indeed, investigated the relationship between institutions and growth using comparative case studies. They have identified certain key institutional functions: protect property, enforce contracts, facilitate financial intermediation, and so on. Even economic historians are sometimes less careful than they could be: The right to property is actually a complex of rights (to earn income from, to use, to change, to sell, to rent, etc.), and financial institutions serve several important functions. The connection between institutions and growth has been little studied in other disciplines, with the important exception of the literature on the role of government in economic development. One area in which much more research is needed across disciplines is identifying the types of institution that serve various institutional functions. Institutions differ a great deal across successful developed countries, and China has experienced rapid growth against a backdrop of very peculiar institutions. Yet, it is clear that there are limits to the range of supportive institutions.

The major insight of the literature on “the developmental state” (primarily in political science, but also sociology) can be briefly summarized: “Sterile debates about ‘how much’ states intervene have to be replaced with arguments about different kinds of involvement and their effects” (Evans, 1995, p. 10). That is, we must move past attempts to identify best institutions to look at how well different institutions are enforced. Scholars of the developmental state (such as the political scientist Kohli, 2004) tend to argue that effective governments are necessary for any growth strategy to be effective. This insight has not been ignored in economics, but in practice it is often neglected: It is generally much easier to identify whether a rule exists than whether it matters. As noted above, the literature on trade and growth in economics has largely eschewed the question of whether countries have the capability to manage whatever trade regime they pursue. And thus, the literature on the developmental state can serve as an important reminder of the importance of institutional quality. Although the connection between institutions and growth has received limited attention in most disciplines, the course of institutional change itself has been investigated across a range of disciplines (interest in institutions was reawakened in many disciplines in the 1980s). This wide range of approaches is potentially complementary, at least once some extreme assumptions are pruned from them:

• Evolutionary theorists might posit that institutional change reflects selection over random institutional changes, while rational choice theorists and functionalists (and social utilitarianism within sociology) might assume that institutions represent the farsighted intentions of their creators. Institutions are likely not the result of either perfect foresight or no foresight at all. So, then, it is an empirical question to what extent agents know what they are doing and whether the institutions created serve the purposes intended. Some scholars within all of these traditions relax assumptions of perfect foresight or no foresight. Game theoretic analyses assume that agents feel their way toward institutional solutions that work. While this is usually done in an equilibrium framework, this element can be relaxed so that institutional change is viewed as a never-finished project.

• Whether equilibrium is assumed or not (but especially if it is not), path dependence—such that small events can have important effects on the results of evolutionary processes—becomes possible and can be used in explaining a variety of types of suboptimal outcomes of historical processes. Path- dependent processes are important not just in economic history but in historical institutionalism in political science, where interactions among a variety of institutions (including influences of political on economic institutions), and among agents of unequal power, generate path-dependent processes. The objection to path dependence comes only from rational choice scholars who assume optimal outcomes.

• Once we move away from assumptions that institutions are purposely designed to serve societal goals, scope is created for a variety of causal arguments. Most obviously, the relative power of different agents comes into play. Power is stressed in sociology and political science and by economists such as Knight or Acemoglu. And economic historians have a long tradition of appreciating the role of power (e.g., in analyses of the feudal system; Greif, 2006). Political scientists tend to tell both good stories of purposeful pursuit of beneficial institutional change and bad stories of the exercise of power; these are likely complementary explanations rather than substitutes.

• The exercise of power is often obscured from view: Those exercising power generally wish not to encourage an angry reaction and thus pretend to have other motives. Analyses of power, then, are entirely compatible with analyses of legitimation. Both sociologists and political scientists speak of legitimation. But as game theory analysis of institutional change suggests, cultural attitudes are not easy to change purposefully. Scholars can usefully investigate, then, the degree to which processes of legitimation serve the interests of the powerful or have a momentum of their own. Social constructionism provides one useful hypothesis here: that institutions solidify over time into a form that comes to seem natural. Another sociological hypothesis is that political institutions themselves shape which other institutions are viewed as legitimate. Both could well capture important aspects of legitimation.

• Sociological treatment of “ideas” encourages a broader exploration of the influence of culture on institutions. If culture is not shaped entirely by the powerful, then cultural values may exert a range of influences on which institutions are thought to be desirable. The normative approach in political science is similar. The “ideas” approach is particularly unique in emphasizing the role that social science might play in establishing the desirability of certain institutions. The seeming reticence of other social scientists to engage this line of argument is remarkable.

• Network analysis in sociology argues that agents are constrained by their networks: Institutional change is only possible if supported actively by a sufficient network. Such an approach needs to look at the resources that different network members bring, and it is thus compatible with an emphasis on power (and culture). The business-labor analyses in political science can be seen as a particular approach to network analysis, looking at how particular coalitions, whether within or across groups, were formed and able to achieve change. Organizational institutionalism in sociology alternatively examines the different motives of agents within an organization.

• Historical institutionalism is not alone in stressing the causal links between institutions. Both structural institutionalism and social institutionalism emphasize the effects of certain sorts of institutions on others. These theories address links to institutions from each of the major categories of phenomena identified in Szostak (2003, 2004), with the exception of the nonhuman environment and the two psychological categories. Because each of these theories tends to be pursued in one or two disciplines, interdisciplinary integration promises a more holistic outlook than any one discipline can provide. The treatment of cultural and social phenomena tends to be too broad and diffuse and needs to more carefully focus on particular cultural and social elements. Beyond this need for greater clarity, there are no obvious missing variables from the set of theories as a whole. In terms of theory types, different theories emphasize individuals (rational choice), relationships (game theory, networks), groups (legitimation, cultural and social arguments), and nonintentional agents (historical institutionalism’s emphasis on interactions among institutions). There are hopeful signs of increased flexibility within theories on this point: Although rational choice theory used to stress individuals and historical institutionalism used to stress groups, both increasingly relate the behaviors of aggregates like unions to those of members (Thelen, 1997, p. 378). Theories of institutional change naturally stress actions (more rarely passive reaction, in some evolutionary and historical approaches), but some note the intermediate role of ideas or values. Rational decision making is mentioned most explicitly, but game theory and evolutionary approaches often refer to an intuitive groping for improvement. Structural institutionalism emphasizes how decision-making processes or rules influence outcomes, but at the level of political institutions rather than individual agents. Organizational institutionalism in sociology stresses the role of routines in behavior. Legitimation approaches have a central place for virtue-based decisions. Path-dependent theories have an obvious place for traditions, though agents need not argue explicitly from tradition in order to generate path- dependent outcomes. Rational choice and game theory analyses stress equilibria (but not necessarily), but most other approaches embrace dynamic or stochastic outcomes. In terms of theory types, then, the major omission is in terms of decision making: More explicit attention to intuitive and traditional, and especially virtue- and rule-based decisions, would be useful. These types of decision making are rarely found in the disciplines that have studied economic growth the most, but they are commonly explored in anthropology and the humanities. In establishing common ground, we would wish to know which theories or causal links are the most important—and whether the relative importance varies by time or place or type of institution. Unfortunately, the relative strengths of different theories are rarely compared in practice, for the simple reason that most researchers favor and master only one.6 The empirical evidence collected in support of each theory is strong enough, arguably, to urge the dismissal of the extreme arguments noted above that only one theory is correct. Legions of political scientists and sociologists and economic historians have been fooling themselves, if assumptions of perfect foresight or functionalism are entirely correct. The debate regarding path dependence is more subtle: Although it is clear that path dependence is important over some time periods, the question of whether optimal institutions are inevitably selected in the end is hard to establish uncontrovertibly, though the diversity of economic institutions in countries of similar levels of development suggests otherwise.

Culture and Growth

It is noteworthy at the outset that the theories of institutional change referred to above almost all had some role for culture. This insight accords with casual empiricism: Laws against littering or drugs are almost impossible to enforce if many members of society view them as illegitimate. Yet, although economists admit the importance of culture in this way, they tend to stop short of explicit cultural analysis. Greif (2006, pp. 8, 19–20) provides a good example of both attitudes. He not only recognizes that we must understand why rules are enforced and obeyed, and cannot thus simply study the development of formal rules in isolation, but goes so far as to define “institution” as a complementary complex of formal rules (what we and most others would call institutions) and cultural elements.7 Institutional change is slow and path dependent because institutionsdepend on “poorly understood and often unintentional processes of socialization, internalization, learning, and experimentation,” including beliefs and ethical attitudes (p. 190). Yet, Greif worries that cultural elements are largely unobservable, and he despairs of cultural explanations of anything for this reason: Because ad hoc appeals to unobservable cultural elements can explain everything, they explain nothing (p. xv). He thus focuses his analysis almost entirely on the observable formal rules, assuming that when these seem to work, supporting cultural values and beliefs must be in place. To argue that culture cannot be operationalized is to suggest that scholars of culture across a range of disciplines and interdisciplinary fields have been wasting their time. Surveys are the most common source of measures of cultural attitudes. Interviews and observation can usefully clarify whether survey questions are both understood and answered honestly. In some cases, indirect quantitative measures are possible: Trust can be defined as whether agents behave as they are expected to. There are thus a variety of ways in which cultural attitudes can be operationalized.8 To be sure, there are problems with the existing scholarship on culture. The term itself is perhaps the vaguest in all of human science: Thousands of different definitions exist, and individual scholars rarely bother to clarify what they mean by the term. Yet, culture can potentially be defined in terms of a host of attitudes and behaviors; moreover, the works of individual cultural scholars can generally be understood as engaging a handful of these (see Szostak, 2003). The ideological content of some versions of cultural studies raises concerns that the important causal role of culture is simply assumed—but similar concerns have been raised regarding other scholarly communities above. The sociologist Beckert (2002) provides a compelling overview of the need for sociocultural analysis of economic decisions. Economic theory itself suggests limits to the exercise of rationality in two common circumstances. The first is when cooperation among agents is required, in which case economic calculations depend on culturally conditioned expectations regarding the behavior of others. The second is when uncertainty is present: If actors cannot rationally attach probabilities to the results that their actions might produce, they must rely on various mental rules to guide behavior. Although individuals will differ in these, there will also be cultural influences (and cooperation in the face of uncertainty would depend on similar mental rules). He urges sociologists to focus on examining the social influences on those economic decisions for which rationality is particularly problematic. This recommendation would have the effect of strengthening the value of the research in each discipline to the other. He also urges sociologists to move away from references to “irrationality” and instead identify specific strategies engaged in by actors when strictly rational calculation is not feasible. This advice is similar to the advice of Newell (2007) to find common ground between economists’ emphasis on rationality and sociologists’ stress on irrationality by thinking of a continuum between the two. Although Beckert (2002) does not attempt to classify these non-strictly rational9 strategies, he makes frequent reference to following routines (which not only reduces costs of calculation, but increases the predictability of the behavior of others) and following cultural guidelines—including respect for widely shared values—and less frequent mention of intuition; thus, his analysis is consistent with the elucidation of the five types of decision making above.

Economists tend to believe that there is one best value set for capitalist societies (Blim 2005, p. 307). Modernization theory in sociology had also suggested that certain values, such as attitudes toward achievement, were essential (p. 308). More recently, Fukuyama has stressed trust and Harrison has urged future orientation, work effort, frugality, education, merit, trust, honesty, justice/fairness, dispersed authority, and secularism (p. 309). Blim devotes several pages to detailing differences in institutions across modern economies and then shows that these both reflect and support value differences. Different emphases on individual versus community are reflected in lifetime employment in Japan and huge CEO salaries in the United States. Values regarding competition versus collaboration are reflected in different approaches to labor/management relations (p. 316). Again, careful specification of causal relationships, in concert with careful attendance to differences across countries, produced a more complicated but more accurate understanding. In sum, the analysis of economists suggests that cultural values may be very important causes of economic growth. Research in other disciplines has rarely addressed growth directly and has too often been characterized by vague terminology and lack of careful empirical analysis. These problems can each be overcome. At present, though, the interdisciplinarian can most usefully provide advice to disciplinarians as to how insights here can best be developed and/or clarified.

Networks

Networks of individuals serve a variety of social, political, and economic purposes. Networks are a third way of organizing economic activity—along with the markets and organizational hierarchies that economists have stressed until recently—and sociologists have shown that networks are critical in such important economic activities as finding a job or locating a business contact. Moreover, network analysis focuses on relationship agency, whereas economic analysis stresses individual agency. Network analysis is thus potentially a valuable addition to the study of economic growth. Economists have rarely used network analysis and tend, when they do, to stress a static analysis of how networks work, rather than looking at how networks evolve. Network analysis has occasionally been applied to economic interactions but not directly to the study of growth. As with culture, then, interdisciplinarians can provide advice on promising research strategies and potential pitfalls. Scholars wishing to examine the relationship between networks and growth could benefit from the following observations:

• Social capital is another vague term. There are three types of definition. The worst invites tautology (and ignores the reality of bad outcomes being possible) by defining social capital solely in terms of results. The second stresses cultural values such as trust. The third emphasizes networks and perhaps organizations. The last two can be combined: Networks generate generally beneficial outcomes by encouraging trust. This chapter will thus not address social capital, but rather address culture and networks separately.

• As with culture, useful classifications of different types of network are needed. Sociologists stress the importance of weak and/or cross-group links: Networks serve a critical role in transmitting information, and the most important links may be the less obvious and less strong links between individuals with access to quite different types of information.

• Then scholars must identify the links between different types of network and different economic activities: investment most obviously, but also innovation (which scholars of both science and technology increasingly appreciate occurs within networks) and institutional change (which is also only possible if resources are mobilized through networks). It should not be forgotten in these explorations that networks can serve to divert resources from growth; links between groups may be particularly important in generating positive outcomes.

• In particular, network analysis may provide a means (though certainly not the only means) to get a handle on the slippery concept of entrepreneurship (another topic treated more often outside economics than within). Entrepreneurs can only be successful within networks, and thus both the supply of entrepreneurs and their effectiveness will be influenced by the availability of networks that provide access to diverse sources of information and resources. Ironically, Swedberg & Granovetter (2001, pp. 12–13) observe that entrepreneurs often only succeed after migrating away from expectations within networks of family and friends (e.g., expectations that they will employ family members).

• The reasons for differences in networks across countries can be explored in a comparative fashion. Granovetter (2001) has argued that there are limits to the ability of an agent to forge links instrumentally for gain. Instead, information often flows as a side-effect of relationships pursued for social reasons. It has been hypothesized that networks are especially weak when social divisions are sharp, poverty exists without a safety net, rule of law is weak, politics is not free and without real choices, different groups do not see shared goals, war or famine undermine sense of stability, and minorities are discriminated against.

• The question of whether networks (or markets) are a substitute or complement for institutions deserves further attention: It is probably true that institutions can substitute for deficient networks in some cases but not others. This question has important policy implications for countries with limited “social capital.”

Social Structure

There is a general appreciation across all disciplines that social divisions—primarily ethnic and class differences, but also gender and family divisions—can have negative economic and political effects. Writers in each discipline show how, in at least some instances, these negative effects are of enormous importance. Despite the general (and unusual) consensus on this point, there are still opportunities for one discipline to learn from another. Economists tend to downplay issues of class or inequality, while other disciplines may exaggerate these issues. Economists rarely look at the sources of social division, and may thus too readily assume that they are intractable. Geddes (2002) warns us not to take ethnic divisions as given, for individuals have choices about how to identify themselves. People also decide whether to feel mistreated or threatened by other groups and how to respond. She argues that this focus on malleability characterizes modern political science research (p. 361).

Community Development

Among disadvantaged groups, both in poor countries and rich countries, economic growth may depend on “community development”: Members of the community need to come together in order to identify strategies that enhance economic growth prospects directly or indirectly (by improving education, health, legal, or a host of other institutions and policies). Local communities often provide critical infrastructure, such as irrigation. The World Bank increasingly makes unrestricted loans to local communities (Stiglitz, 2006, pp. 51–53). Community development involves strengthening civil society (by strengthening links within the community and its interaction with sources of academic and professional advice), in order to prioritize the actions and perspectives of these communities in addressing the development of social/economic/environmental policies. It thus involves empowerment: strengthening the capacity of both individuals and community-level institutions. Community development, like the economic growth it may encourage, is best pursued in an interdisciplinary manner. This is, in part, because community development usually involves complex challenges, and thus it requires input from a variety of academic disciplines and professions. As well, the diversity of insights gives communities a freedom of choice (in how they integrate them) that they lack if presented with only one discipline-grounded policy option. Moreover, there is a synergy between cross-disciplinary integration and the integration of community insights with academic/professional insights: Both types of integration depend on respect, forging a common vocabulary, and seeking a whole that is greater than its parts. Community activists often need to oppose entrenched interests, just as interdisciplinarians must at times confront the entrenched authority of disciplines (see Butterfield & Korazim-Körösy, 2007).

Emergent Properties

We must be careful that the study of individual causal links does not divert our attention from emergent properties that might be apparent only at the level of broader systems of links. Two types of emergent properties deserve particular attention. First, growth itself can be considered to be an emergent property of a host of independent actions, mostly undertaken without having the encouragement of growth as an objective. There is indeed a long tradition in many fields of arguing that economic growth occurs only when many causal forces are combined. “Big push” theories in economic development in the 1960s, Walt Rostow’s “stages of growth” theory with its long list of necessary conditions for a “takeoff into economic growth,” dependency and world systems theory (which postulate a variety of ways in which poor countries might be kept poor by their interactions with rich countries) in sociology and political science, among other approaches, have made this sort of argument. Are the optimistic or pessimistic versions of these approaches more plausible? In terms of the three strategies for dealing with differences in interdisciplinary insights (above), it is clear that this difference is not merely semantic: Pessimists and optimists are talking about the same thing and reaching different conclusions. Do these hypotheses have different ranges of applicability? It must seem that many of the world’s poor countries have achieved impressive rates of economic growth in recent decades and thus better fit the optimistic outlook. Most of sub-Saharan Africa might better fit the pessimist scenario, though again, it must be recalled that some of these countries grew rapidly in the 1960s. There might be some mechanism that ensures that at least some poor countries remain poor. This leads to the third question: Could one achieve common ground by changing some assumptions in one or the other perspective? Indeed one could. If one strips away the determination to identify without doubt the future course of history, each perspective supports and depends upon a set of causal arguments. It is entirely possible that all of these have some empirical merit, and thus whether a country grows or not depends on which causal forces are operating most strongly at certain times. And such a common ground can indeed be seen in the writings of both camps: Optimists often talk about overcoming what they see as surmountable barriers (such as the absence of property rights or decent infrastructure) to economic growth, while pessimists generally suggest that growth will only occur if dramatic changes are made (say, in trade or foreign investment policies at the global level) to the way the world operates

Although both optimists and pessimists appeal to issues of complexity, they have not, in general, emphasized emergent properties but rather a set of mutually reinforcing causal links. This need not mean that emergent properties are unimportant: Scholars across disciplines may be biased toward making narrow causal arguments rather than appealing to emergent properties. A second venue for emergent properties is economic fluctuations. Economic actors do not set out to generate business cycles (indeed, this result is even less intended than economic growth), but cycles emerge from the interactions among actors. Economists have struggled for decades to explain cycles precisely because it is hard to move from an understanding of how individuals behave to an appreciation of how cycles are generated. For present purposes, a few brief points can be made about fluctuations:

• In a world without growth and the structural change that accompanies it, economic cycles would be mild or nonexistent. Cycles should thus be treated as largely an emergent property of the economic growth process.

• In addition to the business cycles of a year or two in duration that economists have focused most attention on, there are longer periods—of a decade or more—characterized by significant differences in growth rates: The 1950s and 1960s saw more rapid growth in most developed countries than did the 1930s or 1970s. Notably, these periods of rapid growth tend also to be characterized by less severe cyclical behavior—perhaps because workers losing their jobs could quickly find others.

• However, economic models of growth predict a steady state outcome of some constant rate of growth. It was noticed above that other disciplines are more likely than economics to stress differences in growth experience across time and place. Interdisciplinarians should thus seek a common ground that reflects the observed reality of alternating periods of fast and slow growth. (Our appreciation above that technological innovation occurs unevenly through time might form an important component of this common ground.) Lessons for Interdisciplinary Practice The foregoing analysis has, I hope, provided evidence of the advantages of a strategy of integrating insights along different causal links (without neglecting the interactions among them) and also seeking emergent properties of the system as a whole. The literature on interdisciplinarity often refers somewhat vaguely to different “facets” of an issue. The stress here on causal links clarifies the focus of analysis and points again to the advantages of an exhaustive table of the key phenomena studied by scholars across all disciplines: This provides a “map” of the links addressed while also mitigating against ignoring links just because all disciplines have ignored them (or the literature survey failed to find works that did). Interdisciplinarians have understandably focused the most on integration when different disciplines provide conflicting insights. Yet, the analysis above suggests that integrative strategies are also useful when gaps exist between disciplines. In such cases, interdisciplinarians can usefully suggest avenues for research that would bridge these gaps. The symbiotic relationship between disciplines and interdisciplinarity may then be

easier to display to disciplinarians: Interdisciplinarians can point to valuable extensions to disciplinary analysis without having first to outline the deficiencies of previous research. Reflection and Communication Establishing common ground is the most important single step in the integrative process. Yet, the interdisciplinarian cannot simply stop at this point. As is true in specialized research, the act of insight should be followed by careful attempts to evaluate and clarify the integrative insights obtained. Interdisciplinarians are not free of bias themselves, though they will generally be more aware of the existence of scholarly biases than specialized researchers. Interdisciplinarians as a whole may be biased toward seeing good (or not) in all approaches, and they must thus be careful to scrutinize each insight they take from any discipline. Individual interdisciplinarians may be characterized by a host of ideological, ethical, epistemological, theoretical, methodological, and other biases. They may like some disciplines more than others. All these possibilities should be reflected upon to see whether the integrative results obtained reflect such biases.10 The interdisciplinarian should then ask whether there are ways in which his or her integrative understanding might be tested. It may well be that a complex integrative understanding such as that sketched above cannot be tested in its entirety (though it is useful to ask whether the set of insights as a whole is useful to the policy maker). Rather, different tests may be required for different causal links. The interdisciplinarian will wish to use multiple methods to test insights. If different methods suggest different conclusions, it is necessary to revisit the strategies outlined above of interrogating assumptions and revisiting the strengths and weaknesses of different theories and methods. The final task involves communicating results in a format that is accessible to multiple audiences. This involves appreciating both the knowledge bases of different audiences and their interests: The results should be connected to issues that different audiences (especially disciplines) already care about.

Conclusion

In the end, did we end up with a chaos of conflicting arguments extending in too many directions? Or were we able to put enough order into our reflections to make the effort worthwhile? Hardcore disciplinarians will respond negatively: They can instill greater order by simply ignoring theories and methods other than their own. But the interdisciplinarian strives toward an order that does not arbitrarily limit insight. The cause of economic growth—and the billions of people who desperately need to experience more of this—is best served by privileging integrated insight over narrower criteria. It is useful to close by briefly reviewing the benefits to our understanding of growth of the various steps in the interdisciplinary research process:

• The scope for interdisciplinary analysis expands markedly when we move beyond studying the proximate causes of growth to ask why some countries innovate or trade or invest more than others. Economists—and far from all of these—have only recently extended their gaze beyond the proximate causes, and they are held back by the difficulty of applying their usual theories and methods to non-economic phenomena such as culture. The importance of not arbitrarily constraining a guiding question along disciplinary lines could hardly be better illustrated.

• The second step(s) reviewed a wide range of theories and methods and established that each had strengths for the study of growth that could compensate for weaknesses in others. The appreciation of these strengths and weaknesses was invaluable when disciplinary insights were evaluated later. The analysis also served usefully to justify in advance wide-ranging theoretical and methodological explorations. For mainstream economists, the key message was that rational choice theorizing and statistical analysis need to be supplemented with other—for the most part complementary—theories and methods. Yet, the same message was communicated to all other disciplines.

• The identification of gaps in scholarly understanding is always a critical step in interdisciplinary analysis. It has been particularly important in the case of economic growth, for scholars of networks or culture or business cycles have only rarely addressed questions of economic growth (and vice veser), and even much of the literature on technology and institutions is oriented toward quite different questions. One of the main purposes of this research project has been to identify areas where future research is needed and how this might be pursued.

• Asking about the possibility of emergent properties led us to two important areas of investigation: poverty traps and the connection between growth and cycles.

• More generally, we were able to create a common ground across a range of causal links that is superior to the insights of any one discipline. We will not reprise those analyses here, in part not to detract attention from the other steps in the research process.

• Though we did not have space to describe the final steps in detail, our understanding of growth will be greatly enhanced if we interrogate possible biases in our analysis, test our insights empirically, and communicate them clearly to diverse audiences.

1. This chapter draws on Szostak (2009). That book is organized according to the 12 steps identified in Szostak (2002). In this chapter, they are combined into four sets of steps. My process is broadly similar to that in Repko (2008), though I tend to stress the identification of relevant phenomena, theories, and methods more than Repko, and I also rely more than Repko on classifications of phenomena, theories, and methods in evaluating disciplinary insights. Yet these are differences in emphasis only; Repko and I concur regarding the broad outlines of the research process. In the book, I address the question of whether growth is “good.” For rich countries especially, I urge the definition and measurement of growth in terms of output per hours worked (and thus growth might just mean more leisure time) and with due regard to environmental repercussions. I also stress that only some goods and services add to human well-being.

2. These are the three possible types of “intentional agency” or “nonintentional agency” identified in Szostak (2004). They together form one of the five dimensions in the typology of theory cited in Repko (2008, pp. 198–200). Others are addressed below. Durkheim had distinguished sociology from economics and psychology by emphasis on methodological holism. “Methodological holism in sociology has been an obstacle to acceptance of the choice-theoretic approach underlying the new institutional paradigm”—it has isolated sociology from changes in other social sciences (Nee, 1998, p. 11). But while this approach has dominated, many analyses, from Tocqueville and Weber to today, have emphasized “rational action bounded by institutions” (p. 4). Rather than debate individualism versus holism, “a more constructive approach is to model the reciprocal interaction between purposive action and social structure” (p. 5). In other words, these approaches can be integrated.

3. Some scholars would worry about the number of distinct causal links that would need to be investigated in any complex study. We can hardly hope, though, for a simple understanding of a complex process. Szostak (2009) shows how the causal link understandings can be organized into a coherent whole through reference to an exhaustive classification of phenomena.

4. Economists, as we shall see, worry a great deal about the quality of institutions when discussing institutions in general. That is, they recognize that countries differ a great deal in how well they manage/enforce institutions that may look quite similar on paper. In the realm of trade policy, however, analysis tends to proceed with respect to a dichotomy between openness and managed trade. There is usually an implicit or explicit assumption that countries cannot manage trade very well. Kohli (2004), a political scientist, argues that the key difference in developmental prospects is between countries that can manage/enforce any institutions well and countries that can manage/enforce no economic institutions well. He argues, for example, that South Korea effectively managed several years of import substitution as well as decades of export promotion (i.e., the government was not “captured” by private industries and encouraged them to improve productivity under both regimes). On the other hand, Nigeria failed miserably with respect to both types of policies (both were perverted to reward friends of the government, and productivity advance was not encouraged; p. 376). The broader literature on “the developmental state” makes similar arguments.

5. Historians of technology give roughly equal attention to the technological, economic, political, and sociocultural influences on innovation. They increasingly pay attention to the long period of development after a breakthrough innovation. They note that as technological systems harden, they limit human choices about technological innovation (Nye, 2006).

6. The value of integrating these approaches has occasionally been appreciated in the literature. Thelen (1997, p. 370) suggests that rational choice theorists can appreciate historical circumstances while historical institutionalists can think more about why actors do what they do (and the importance of collective action problems); all can recognize the role of norms in supporting institutions. Thelen’s main argument is that scholars cannot understand change without also understanding stability. Both the coordination emphasized by rational choice theory and the shared cultural understandings of sociological institutionalism allow us to understand continuity better than change (p. 386). Historical institutionalism, on the other hand, invokes sunk costs and vested interests, and thus is good at identifying critical junctures that send countries on different trajectories, but worse at explaining continuity.

7. Greif (2006) notes that different disciplines define institutions differently: as rules, as norms, as functional solutions to particular problems. He urges the integration of these definitions. From our holistic perspective, this is best done not by conflating quite distinct phenomena but by capturing in different causal links both the influence of culture on institutions and the role of institutions in solving particular problems. Greif’s own wish that different definitions be viewed as complements (p. 40) is best achieved in this manner. Greif’s definition encourages an emphasis on culture over other causal influences on/of institutions, while inviting us to treat the culture-institutions nexus as a black box.

8. In the past decades, the availability of better data and techniques has induced some economists to study culture. One popular approach is to look at whether membership in a particular ethnic or religious group affects economic outcomes. The advantage of this approach is that such memberships are easily measured and are also largely inherited: This overcomes the possible concern that any correlation between culture and economy reflects causation in the other direction. Both internationally and within countries, ethnic and religious differences do generate different economic outcomes (though within countries, studies of immigrants suggest that these will lessen over time). Moreover, these differences are correlated with different values and beliefs (trust, social mobility, fairness, hard work, fertility, thrift) in both regressions and experiments. Yet, such studies can only be suggestive of links between these values and economic growth, and these links are hard to establish statistically (Guiso, Spienza, & Zingales, 2006).

9. Beckert (2002), in distinguishing himself from the emphasis of other sociologists on irrationality, strives to emphasize the “rationality” of other decision-making strategies. In this chapter, these can be seen as reasoned but nonrational strategies. Note that semantic confusion between narrow and broad uses of the word “rational” contribute to misunderstanding between sociologists and economists.

10. I am an economist, but one who has written methodological critiques of my discipline and my field of economic history (Szostak, 1999, 2006). I have thus had the pleasure of being critiqued by non-economists for being too much of an economist and by economists for being not quite enough of one. The reader can best judge which—likely both—is the case here. I lack practitioner-level expertise in some of the disciplines covered here, but I have considerable familiarity with most of the theory types and methods addressed. I am a self-conscious interdisciplinarian and thus likely biased toward stressing the advantages of interdisciplinary analysis. As should be clear by now, I believe in theoretical and methodological flexibility. I may thus be biased toward seeing some good in all approaches. Indeed, I do suspect that any idea pursued at length by some academic community must have some kernel of truth in it. But this need not prevent skepticism: One can appreciate that those who thought the world was flat for millennia were misguided, while appreciating the value of the ways they amassed evidence in support of their hypothesis. Still, I can imagine that disciplinarians reading this chapter will readily imagine that I have been too harsh with respect to them and not harsh enough in my treatment of others. And surely some of them will be right (but hopefully not to a considerable degree), though I know not which.

References

Aron, J. (2000). Growth and institutions: A review of the evidence. World Bank Research Observer, 15(1), 99–135. http://dx.doi.org/10.1093/wbro/15.1.99 Bammer, G. (2005). Integration and implementation sciences: Building a new specialization. Ecology and http://dx.doi.org/10.1093/wbro/15.1.99

Society, 10(2), 6. Retrieved from http://www.ecologyandsociety.org/vol10/iss2/art6/ Beckert, J. (2002). Beyond the market: The social foundations of economic efficiency. Princeton: Princeton University Press.

Blim, M. (2005). Culture and economy. In Edited by: J. G.Carrier (Ed.), A handbook of economic anthropology (pp. 306–322). Cheltenham: Edward Elgar.

Butterfield, A. K. J., & Korazim-Körösy, Y. (2007). Interdisciplinary community development: International perspectives. New York: Haworth Press.

Cohen, J. D. (2005). The vulcanization of the human brain: A neural perspective on interactions between cognition and emotion. Journal of Economic Perspectives, 19(4), 3–24. http://dx.doi.org/10.1257/ 089533005775196750 Easterly, W. (2002). The elusive quest for growth: Economist’s adventures and misadventures in the tropics.

Cambridge: MIT Press.

Evans, P. (1995). Embedded autonomy: States and industrial transformation. Princeton: Princeton University Press.

Geddes, B. (2002). The great transformation in the study of politics in developing countries. In Edited by:

I.Katznelson & H. V.Miller (Eds.), Political science: State of the discipline (pp. 342–370). New York: Norton.

Gerschenkron, A. (1962). Economic backwardness in historical perspective. Cambridge: Harvard University Press.

Granovetter, M. (2001). Economic action and social structure: The problem of embeddedness. In Edited by:

M.Granovetter & R.Swedberg (Eds.), The sociology of economic life (pp. 51–76). Boulder CO: Westview Press.

Greif, A. (2006). Institutions and the path to the modern economy: Lessons from medieval trade. New York:

Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511791307

Guiso, L., Spienza, P., & Zingales, L. (2006). Does culture affect economic outcomes? Journal of Economic Perspectives, 20(2), 23–48. http://dx.doi.org/10.1257/jep.20.2.23

Kohli, A. (2004). State directed development: Political power and industrialization in the global periphery.

New York: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511754371 Lal, D. (1988). Unintended consequences: The impact of factor endowments, culture, and politics on long-run economic performance. Cambridge: MIT Press.

Meeus, M. T. H., & Hage, J. (2006). Product and process innovation, scientific research, knowledge

http://www.ecologyandsociety.org/vol10/iss2/art6/

http://dx.doi.org/10.1257/089533005775196750

http://dx.doi.org/10.1017/CBO9780511791307

http://dx.doi.org/10.1257/jep.20.2.23

http://dx.doi.org/10.1017/CBO9780511754371

dynamics, and institutional change: An introduction. In Edited by: J.Hage & M. T. H.Meeus (Eds.), Innovation,

science, and institutional change: A research handbook (pp. 1–19). Oxford: Oxford University Press.

Miller, R. (2008). International political economy. London: Routledge. http://dx.doi.org/10.4324/ 9780203927236 Nee, V. (1998). Sources of the new institutionalism. In Edited by: M. C.Brinton & V.Nee (Eds.), The new institutionalism in sociology (pp. 1–16). New York: Russell Sage Foundation.

Newell, W. H. (2007). Decision making in interdisciplinary studies. In Edited by: G.Morcol (Ed.), Handbook of decision making (pp. 245–264). New York: Marcel Dekker.

Nye, D. E. (2006). Technology matters: Questions to live with. Cambridge: MIT Press.

Repko, A. (2008). Interdisciplinary research: Process and theory. Thousand Oaks, CA: Sage.

Singleton, R. A., Jr., & Strait, B. C. (1999). Approaches to social research (3rd. ed.). New York: Oxford University Press.

Snowdon, B. (2002). Conversations on growth, stability, and trade: A historical perspective. Cheltenham: Edward Elgar.

Solow, R. (2005). Reflections on growth theory. In Edited by: P.Aghion & S. N.Durlauf (Eds.), Handbook of economic growth (pp. 1–10). Amsterdam: Elsevier North-Holland.

Stiglitz, J. (2006). Making globalization work. New York: Norton.

Swedberg, R., & Granovetter, M. (2001). Introduction to the second edition. In Edited by: M.Granovetter & R.Swedberg (Eds.), The sociology of economic life (pp. 1–28). Boulder CO: Westview Press.

Edited by: Sylla, R., & Toniolo, G. (Eds.). (1991). Patterns of European industrialization: The nineteenth century. London: Routledge.

Szostak, R. (1999). Econ-art: Divorcing art from science in modern economics. London: Pluto Press.

Szostak, R. (2002). How to do interdisciplinarity: Integrating the debate. Issues in Integrative Studies, 20, 103–122.

Szostak, R. (2003). A schema for unifying human science: Interdisciplinary perspectives on culture.

Selinsgrove PA: Susquehanna University Press.

Szostak, R. (2004). Classifying science: Phenomena, data, theory, method, practice. Dordrecht: Springer.

Szostak, R. (2006). Economic history as it is and should be: Toward an open, honest, methodologically flexible, theoretically diverse, interdisciplinary exploration of the causes and consequences of economic http://dx.doi.org/10.4324/9780203927236 http://dx.doi.org/10.4324/9780203927236 growth. Journal of Socio-Economics, 35(4), 727–750. http://dx.doi.org/10.1016/j.socec.2005.11.006 Szostak, R. (2007). Interdisciplinarity and the classification of scholarly documents by phenomena, theories, and methods. In Edited by: B. RodriguezBravo & L. AlviteDiez (Eds.), Interdisciplinarity and transdisciplinarity in the organization of scientific knowledge: Actas del VIII Congreso ISKO-Espana (pp.

469–477). Leon: University of Leon.

Szostak, R. (2008). Classification, interdisciplinarity, and the study of science. Journal of Documentation, 64(3), 319–332. http://dx.doi.org/10.1108/00220410810867551 Szostak, R. (2009). The causes of economic growth: Interdisciplinary perspectives. Berlin: Springer.

http://dx.doi.org/10.1007/978-3-540-92282-7 Thelen, K. (1997). Historical institutionalism in comparative politics. Annual Review of Political Science, 2, 369–404. http://dx.doi.org/10.1146/annurev.polisci.2.1.369 http://dx.doi.org/10.4135/9781483349541.n6

http://dx.doi.org/10.1016/j.socec.2005.11.006

http://dx.doi.org/10.1108/00220410810867551

http://dx.doi.org/10.1007/978-3-540-92282-7

http://dx.doi.org/10.1146/annurev.polisci.2.1.369

http://dx.doi.org/10.4135/9781483349541.n6

An Interdisciplinary Analysis of the Causes of Economic Growth

In: Case Studies in Interdisciplinary Research

Collepals.com Plagiarism Free Papers

Are you looking for custom essay writing service or even dissertation writing services? Just request for our write my paper service, and we'll match you with the best essay writer in your subject! With an exceptional team of professional academic experts in a wide range of subjects, we can guarantee you an unrivaled quality of custom-written papers.

Get ZERO PLAGIARISM, HUMAN WRITTEN ESSAYS

Why Hire Collepals.com writers to do your paper?

Quality- We are experienced and have access to ample research materials.

We write plagiarism Free Content

Confidential- We never share or sell your personal information to third parties.

Support-Chat with us today! We are always waiting to answer all your questions.

Can economic growth be understood by relying only on the insights of one discipline? Or does our understanding increase markedly if insights from many disciplines are integrated?

Related Posts

John Doe has decided to clone himself. He is sterile

Prepare a 2- to 3-page paper (not counting the cover and reference pages) discussing how the Family and Medical Leave Act, the Pregnancy Discrimination Act, the Americans with Disabilities Act, safety laws, or other state laws might apply to the following scenario.

Capital Budget Methodologies