Arnold Kling (2016), Specialization and Trade, A Re-Introduction to Economics, Washington, DC, Cato Institute.

Kling’s central theme in this short book of nine main chapters is that specialization, trade, and the coordination of individual plans by means of the price system and the profit motive play fundamental roles in modern economies. Most mainstream economists would agree with this assessment. Their models of trade, growth, and innovation certainly include the four elements, with varying emphasis.

But Kling criticizes the methodological approach adopted by post-world-war-II economics, which he associates with “MIT economics.” An MIT PhD himself, he argues that economics, and specifically macroeconomics, should adopt less of a mechanistic and more of an evolutionary perspective to gain relevance. In the second chapter, entitled “Machine as Metaphor,” Kling asserts that under the leadership of Paul Samuelson post-war (macro)economics framed economic issues as programming problems that resemble resource allocation problems in a wartime economy. Even as the discipline evolved, Kling contends, the methodology remained the same, pretending controllability by economist-engineers; in the process, the role of specialization was sidelined in the analysis.

I think that Kling is too harsh in his assessment. Economics and macroeconomics, in particular, has changed dramatically since the times of Paul Samuelson. The notion that, given enough instruments, any economic problem can be solved as easily as a system of equations, has lost attraction. Modern macroeconomic models are based on microeconomic primitives; they take gains from trade seriously; they involve expectations and frictions; and they do not suggest easy answers. The task of modern macroeconomics is not to spit out a roadmap for the economist-engineer but to understand mechanisms and identify problems that arise from misaligned incentives.

Kling is right, of course, when he argues that many theoretical models are too simplistic to be taken at face value. But this is not a critique against economic research which must focus and abstract in order to clarify. It rather is a critique against professional policy advisors and forecasters, “economic experts” say. These “experts” face the difficult task of surveying the vast variety of mechanisms identified by academic research and to apply judgement when weighing their relevance for a particular real-world setting. To be useful, “experts” must not rely on a single framework and extrapolation. Instead, they must base their analysis on a wide set of frameworks to gain independent perspectives on a question of interest.

In chapters three to five, Kling discusses in more detail the interplay of myriads of specialized trading partners in a market economy and how prices and the profit motive orchestrate it. In the chapter entitled “Instructions and Incentives,” Kling emphasizes that prices signal scarcity and opportunity costs are subjective. In the chapter entitled “Choices and Commands,” he discusses that command-and-control approaches to organizing a society face information, incentive, and innovation problems, unlike approaches that rely on a functioning price mechanism. And in the chapter “Specialization and Sustainability,” Kling makes the point that well-defined property rights and a functioning price mechanism offer the best possible protection for scarce resources and a guarantee for their efficient use. Sustainability additionally requires mechanisms to secure intergenerational equity.

I agree with Kling’s point that we should be humble when assessing whether market prices, which reflect the interplay of countless actors, are “right” or “wrong.” However, I would probably be prepared more often than Kling to acknowledge market failures of the type that call for corrective taxes. The general point is that Kling’s views expressed in the three chapters seem entirely mainstream. While we may debate how often and strongly market prices fail to account for social costs and benefits, the economics profession widely agrees that for a price system to function well this precondition must be satisfied.

In the sixth chapter, entitled “Trade and Trust,” Kling argues that specialization rests on cultural evolution and learning and more broadly, that modern economic systems require institutions that promote trust. Independently of the norms a particular society adopts, it must implement the basic social rule,

[r]eward cooperators and punish defectors.

How this is achieved (even if it is against the short-run interest of an individual) varies. Incentive mechanisms may be built on the rule of law, religion, or reputation. And as Kling points out societies almost always rely on some form of government to implement the basic social rule. In turn, this creates problems of abuse of power as well as “deception” and “demonization.” Mainstream economists would agree. In fact, incentive and participation constraints, lack of commitment, enforcement, and self-enforcement are at center stage in many of their models of partial or general equilibrium. Similarly, the role of government, whether benevolent or representing the interests of lobby groups and elites, is a key theme in modern economics.

Chapter seven, entitled “Finance and Fluctuations,” deals with the role of the financial sector. Kling argues that finance is a key prerequisite for specialization and since trust is a prerequisite for finance, swings in trust—waves of optimism and pessimism—affect the economy. No mainstream macroeconomist will object to the notion that the financial sector can amplify shocks. Seminal articles (which all were published well before the most recent financial crisis) exactly make that point. But Kling is probably right that the profession’s workhorse models have not yet been able to incorporate moods, fads, and manias, the reputation of intermediaries, and the confidence of their clients in satisfactory and tractable ways, in spite of recent path-breaking work on the role of heterogenous beliefs.

In chapter eight, Kling focuses on “Policy in Practice.” He explains why identifying market failure in a model is not the same as convincingly arguing for government intervention, simply because first, the model may be wrong and second, there is no reason to expect government intervention to be frictionless. I don’t know any well-trained academic economist who would disagree with this assessment (but many “experts” who are very frighteningly confident about their level of understanding). The profession is well aware of the insights from Public Choice and Political Economics, although these insights might not be as widely taught as they deserve. And Kling is right that economists could explain better why real-world policy selection and implementation can give rise to new problems rather than solely focusing on the issue of how an ideal policy might improve outcomes.

To me, the most interesting chapters of the book are the first and the last, entitled “Filling in Frameworks” and “Macroeconomics and Misgivings,” respectively. In the first chapter, Kling discusses the difference between the natural sciences and economics. He distinguishes between scientific propositions, which a logical flaw or a contradictory experiment falsifies, and “interpretive frameworks” a.k.a. Kuhn’s paradigms, which cannot easily be falsified. Kling argues that

[i]n natural science, there are relatively many falsifiable propositions and relatively few attractive interpretive frameworks. In the social sciences, there are relatively many attractive interpretive frameworks and relatively few falsifiable propositions.

According to Kling, economic models are interpretative frameworks, not scientific propositions, because they incorporate a plethora of auxiliary assumptions and since experiments of the type run in the natural sciences are beyond reach in the social sciences. Anomalies or puzzles do not lead economists to reject their models right away as long as the latter remain useful paradigms to work with. And rightly so, according to Kling: For an interpretative framework with all its anomalies is less flawed than intuition which is uninformed by a framework. At the same time, economists should remain humble, acknowledge the risk of confirmation bias, and remain open to competing interpretative frameworks.

In the chapter entitled “Macroeconomics and Misgivings,” Kling criticizes macroeconomists’ reliance on models with a representative agent. I agree that representative agent models are irrelevant for applied questions when the model implications strongly depend on the assumption that households are literally alike, or that markets are complete such that heterogeneous agents can perfectly insure each other. When “experts” forecast macroeconomic outcomes based on models with a homogeneous household sector then these forecasts rest on very heroic assumptions, as any well-trained economist will readily acknowledge. Is this a problem for macroeconomics which, by the way, has made a lot of progress in modeling economies with heterogeneous agents and incomplete markets? I don’t think so. But it is a problem when “experts” use such inadequate models for policy advice.

Kling argues that the dynamic process of creative destruction that characterizes modern economies requires ongoing change in the patterns of specialization and trade and that this generates unemployment. Mainstream models of innovation and growth capture this process, at least partially; they explain how investment in new types of capital and “ideas” can generate growth and structural change. And the standard framework for modeling labor markets features churn and unemployment (as well as search and matching) although, admittedly, it does not contain a detailed description of the sources of churn. The difference between the mainstream’s and Kling’s view of how the macroeconomy operates thus appears to be a difference of degree rather than substance. And the difference between these views and existing models clearly also reflects the fact that modeling creative destruction and its consequences is difficult.

Kling is a sharp observer when he talks about the difference between “popular Keynesianism” and “rigor-seeking Keynesianism.” The former is what underlies the thinking of many policy makers, central bankers, or journalists: a blend of the aggregate-demand logic taught to undergraduates and some supply side elements. The latter is a tractable simplification of a micro-founded dynamic general equilibrium model with frictions whose properties resemble some key intuitions from popular Keynesianism.

The two forms of Keynesianism help support each other. Popular Keynesianism is useful for trying to convince the public that macroeconomists understand macroeconomic fluctuations and how to control them. Rigor-seeking Keynesianism is used to beat back objections raised by economists who are concerned with the ways in which Keynesianism deviates from standard economics, even though the internal obsessions of rigor-seeking Keynesianism have no traction with those making economic policy.

There is truth to this. But in my view, this critique does not undermine the academic, rigor-seeking type of Keynesianism while it should undermine our trust in “experts” who work with the popular sort which, as Kling explains, mostly is confusing for a trained economist.

In the end, Kling concludes that it is the basics that matter most:

[B]etter economic outcomes arise when patterns of sustainable specialization and trade are formed. … It requires the creative, decentralized, trial-and-error efforts of thousands of entrepreneurs and millions of households … Probably the best thing that the government can do to encourage new forms of specialization is to rethink existing policies that restrict competition, discourage innovation, and retard mobility.

This is a reasonable conclusion. But it is neither a falsifiable proposition nor an interpretive framework. It is the synthesis of many interpretive frameworks, weighed by Kling. In my own view, the weighting is based on too harsh an assessment according to which many modern macroeconomic models are irrelevant.

Kling’s criticism of contemporaneous macroeconomics reads like a criticism of the kind of macroeconomics still taught at the undergraduate level. But modern macroeconomics has moved on—it is general equilibrium microeconomics. Its primary objective is not to produce the one and only model for economist-engineers or “experts” to use, but rather to help us understand mechanisms. A good expert knows many models, is informed about institutions, and has the courage to judge which of the models (or mechanisms they identify) are the most relevant in a specific context. We don’t need a new macroeconomics. But maybe we need better “experts.”