You are looking at 1-20 of 41 articles
Traditional historiography has overestimated the significance of long-distance trade in the medieval economy. However, it could be argued that, because of its dynamic nature, long-distance trade played a more important role in economic development than its relative size would suggest. The term commercial revolution was introduced in the 1950s to refer to the rapid growth of European trade from about the 10th century. Long-distance trade then expanded, with the commercial integration of the two economic poles in the Mediterranean and in Flanders and the contiguous areas. It has been quantitatively shown that the integration of European markets began in the late medieval period, with rapid advancement beginning in the 16th century.
The expansion of medieval trade has been attributed to advanced business techniques, such as the appearance of new forms of partnerships and novel financial and insurance systems. Many economic historians have also emphasized merchants’ relations, especially the establishment of networks to organize trade. More recently, major contributions to institutional economic history have focused on various economic institutions that reduced the uncertainties inherent in premodern economies.
The early reputation-based institutions identified in the literature, such as the systems of the Maghribis in the Mediterranean, Champagne fairs in France, and the Italian city-states, were not optimal for changing conditions that accompanied expansion of trade, as the number of merchants increased and the relations among them became more anonymous, as generally happened during the Middle Ages. An intercommunal conciliation mechanism evolved in medieval northern Europe that supported trade among a large number of distant communities. This institution encouraged merchants to travel to distant towns and establish relations, even with persons they did not already know.
While it is a long-standing idea in international macroeconomic theory that flexible nominal exchange rates have the potential to facilitate adjustment in international relative prices, a monetary union necessarily forgoes this mechanism for facilitating macroeconomic adjustment among its regions. Twenty years of experience in the eurozone monetary union, including the eurozone crisis, have spurred new macroeconomic research on the costs of giving up nominal exchange rates as a tool of adjustment, and the possibility of alternative policies to promote macroeconomic adjustment. Empirical evidence paints a mixed picture regarding the usefulness of nominal exchange rate flexibility: In many historical settings, flexible nominal exchanges rates tend to create more relative price distortions than they have helped resolve; yet, in some contexts exchange rate devaluations can serve as a useful correction to severe relative price misalignments.
Theoretical advances in studying open economy models either support the usefulness of exchange rate movements or find them irrelevant, depending on the specific characteristics of the model economy, including the particular specification of nominal rigidities, international openness in goods markets, and international financial integration. Yet in models that embody certain key aspects of the countries suffering the brunt of the eurozone crisis, such as over-borrowing and persistently high wages, it is found that nominal devaluation can be useful to prevent the type of excessive rise in unemployment observed.
This theoretical research also raises alternative polices and mechanisms to substitute for nominal exchange rate adjustment. These policies include the standard fiscal tools of optimal currency area theory but also extend to a broader set of tools including import tariffs, export subsidies, and prudential taxes on capital flows. Certain combinations of these policies, labeled a “fiscal devaluation,” have been found in theory to replicate the effects of a currency devaluation in the context of a monetary union such as the eurozone. These theoretical developments are helpful for understanding the history of experiences in the eurozone, such as the eurozone crisis. They are also helpful for thinking about options for preventing such crises in the future.
The Hou–Xue–Zhang q-factor model says that the expected return of an asset in excess of the risk-free rate is described by its sensitivities to the market factor, a size factor, an investment factor, and a return on equity (ROE) factor. Empirically, the q-factor model shows strong explanatory power and largely summarizes the cross-section of average stock returns. Most important, it fully subsumes the Fama–French 6-factor model in head-to-head spanning tests.
The q-factor model is an empirical implementation of the investment-based capital asset pricing model (the Investment CAPM). The basic philosophy is to price risky assets from the perspective of their suppliers (firms), as opposed to their buyers (investors). Mathematically, the investment CAPM is a restatement of the net present value (NPV) rule in corporate finance. Intuitively, high investment relative to low expected profitability must imply low costs of capital, and low investment relative to high expected profitability must imply high costs of capital. In a multiperiod framework, if investment is high next period, the present value of cash flows from next period onward must be high. Consisting mostly of this next period present value, the benefits to investment this period must also be high. As such, high investment next period relative to current investment (high expected investment growth) must imply high costs of capital (to keep current investment low).
As a disruptive innovation, the investment CAPM has broad-ranging implications for academic finance and asset management practice. First, the consumption CAPM, of which the classic Sharpe–Lintner CAPM is a special case, is conceptually incomplete. The crux is that it blindly focuses on the demand of risky assets, while abstracting from the supply altogether. Alas, anomalies are primarily relations between firm characteristics and expected returns. By focusing on the supply, the investment CAPM is the missing piece of equilibrium asset pricing. Second, the investment CAPM retains efficient markets, with cross-sectionally varying expected returns, depending on firms’ investment, profitability, and expected growth. As such, capital markets follow standard economic principles, in sharp contrast to the teachings of behavioral finance. Finally, the investment CAPM validates Graham and Dodd’s security analysis on equilibrium grounds, within efficient markets.
Lawrence J. Lau
Chinese real gross domestic product (GDP) grew from US$369 billion in 1978 to US$12.7 trillion in 2017 (in 2017 prices and exchange rate), at almost 10% per annum, making the country the second largest economy in the world, just behind the United States. During the same period, Chinese real GDP per capita grew from US$383 to US$9,137 (2017 prices), at 8.1% per annum.
Chinese economic reform, which began in 1978, consists of two elements—introduction of free markets for goods and services, coupled with conditional producer autonomy, and opening to international trade and direct investment with the rest of the world. In its transition from a centrally planned to a market economy, China employed a “dual-track” approach—with the pre-existing mandatory central plan continuing in force and the establishment of free markets in parallel. In its opening to the world, China set a competitive exchange rate for its currency, made it current account convertible in 1994, and acceded to the World Trade Organisation (WTO) in 2001. In 2005, China became the second largest trading nation in the world, after the United States. Other Chinese policies complementary to its economic reform include the pre-existing low non-agricultural wage and the limit of one-child per couple, introduced in 1979 and phased out in 2016.
The high rate of growth of Chinese real output since 1978 can be largely explained by the high rates of growth of inputs, but there were also other factors at work. Chinese economic growth since 1978 may be attributed as follows: (a) the elimination of the initial economic inefficiency (12.7%), (b) the growth of tangible capital (55.7%) and labor (9.7%) inputs, (c) technical progress (or growth of total factor productivity (TFP)) (8%), and (d) economies of scale (14%).
The Chinese economy also shares many commonalities with other East Asian economies in terms of their development experiences: the lack of natural endowments, the initial conditions (the low real GDP per capita and the existence of surplus agricultural labor), the cultural characteristics (thrift, industry, and high value for education), the economic policies (competitive exchange rate, export promotion, investment in basic infrastructure, and maintenance of macroeconomic stability), and the consistency, predictability, and stability resulting from continuous one-party rule.
Roger E. A. Farmer
The indeterminacy school in macroeconomics exploits the fact that macroeconomic models often display multiple equilibria to understand real-world phenomena. There are two distinct phases in the evolution of its history. The first phase began as a research agenda at the University of Pennsylvania in the United States and at CEPREMAP in Paris in the early 1980s. This phase used models of dynamic indeterminacy to explain how shocks to beliefs can temporarily influence economic outcomes. The second phase was developed at the University of California Los Angeles in the 2000s. This phase used models of incomplete factor markets to explain how shocks to beliefs can permanently influence economic outcomes. The first phase of the indeterminacy school has been used to explain volatility in financial markets. The second phase of the indeterminacy school has been used to explain periods of high persistent unemployment. The two phases of the indeterminacy school provide a microeconomic foundation for Keynes’ general theory that does not rely on the assumption that prices and wages are sticky.
Italy played a central role in the Euro-Mediterranean economy during Antiquity, the late Middle Ages, and the Renaissance. Until the end of the 16th century, the Italian economy was relatively advanced compared with those of the Western European and Mediterranean countries. From the 17th century until the end of the 19th, GDP rose as the population increased. Yet per capita income slowly diminished together with real wages, urbanization, and living standards. Italy lost its central position in the Euro-Mediterranean world and, until the end of the 19th century, was a relatively backward area on the periphery of the most dynamic countries in the north and center of Europe. The Italian premodern economy represents a classic example of extensive growth or GDP growth without improvement in per capita income and living standards.
Hites Ahir and Prakash Loungani
On average across countries, house prices have been on an upward trend over the past 50 years, following a 100-year period over which there was no long-term increase. The rising trend in prices reflects a demand boost due to greater availability of housing finance running up against supply constraints, as land has increasingly become a fixed factor for many reasons. The entire 150-year period has been marked by boom and bust cycles around the trend. These also reflect episodes of demand momentum—due to cheap finance or reasonable or unreasonable expectations of higher incomes—meeting a sluggish supply response. Policy options to manage boom–bust cycles, given the significant costs to the economy from house price busts, are discussed.
Scott Baier and Samuel Standaert
The gravity model of international trade states that the volume of trade between two countries is proportional to their economic mass and a measure of their relative trade frictions. Perhaps because of its intuitive appeal, the gravity model has been the workhorse model of international trade for more than 50 years. While the initial empirical work using the gravity model lacked sound theoretical underpinnings, the theoretical developments have highlighted how a gravity-like specification can be derived from many models with varying assumptions about preferences, technology, and market structure. Along the strengthening of the theoretical roots of the gravity model, the way in which it is estimated has also evolved significantly since the start of the new millennium. Depending on the exact characteristics of regression, different estimation methods should be used to estimate the gravity model.
Leandro Prados de la Escosura and Blanca Sánchez-Alonso
In assessments of modern-day Spain’s economic progress and living standards, inadequate natural resources, inefficient institutions, lack of education and entrepreneurship, and foreign dependency are frequently blamed on poor performance up to the mid-20th century, but no persuasive arguments were provided to explain why such adverse circumstances reversed, giving way to the fast transformation that started in the 1950s. Hence, it is necessary to first inquire how much economic progress has been achieved in Spain and what impact it had on living standards and income distribution since the end of the Peninsular War to the present day, and second to provide an interpretation.
Research published in the 2010s supports the view that income per person has improved remarkably, driven by increases in labor productivity, which derived, in turn, from a more intense and efficient use of physical and human capital per worker. Exposure to international competition represented a decisive element behind growth performance. From an European perspective, Spain underperformed until 1950. Thereafter, Spain’s economy managed to catch up with more advanced countries until 2007. Although the distribution of the fruits of growth did not follow a linear trend, but a Kuznetsian inverted U pattern, higher levels of income per capita are matched by lower inequality, suggesting that Spaniards’ material wellbeing improved substantially during the modern era.
Stephen F. Diamond
Insider trading is not widely understood. Insiders of corporations can, in fact, buy and sell shares of those corporations. But, over time, Congress, the courts and the Securities and Exchange Commission (SEC) have imposed significant limits on such trading. The limits are not always clearly marked and the principles underlying them not always consistent. The core principle is that it is illegal to trade if one is in the possession of material, nonpublic information. But the rationality of this principle has been challenged by successive generations of law and economics scholars, most notably Manne, Easterbrook, Epstein, and Bainbridge. Their “economic” analysis of this contested area of the law provides, arguably, at least a more consistent basis upon which to decide when trades by insiders should, in fact, be disallowed. A return to genuine “first principles” generated by the nature of capitalism, however, allows for more powerful insights into the phenomenon and could lead to more effective regulation.
Henrik Cronqvist and Désirée-Jessica Pély
Corporate finance is about understanding the determinants and consequences of the investment and financing policies of corporations. In a standard neoclassical profit maximization framework, rational agents, that is, managers, make corporate finance decisions on behalf of rational principals, that is, shareholders. Over the past two decades, there has been a rapidly growing interest in augmenting standard finance frameworks with novel insights from cognitive psychology, and more recently, social psychology and sociology. This emerging subfield in finance research has been dubbed behavioral corporate finance, which differentiates between rational and behavioral agents and principals.
The presence of behavioral shareholders, that is, principals, may lead to market timing and catering behavior by rational managers. Such managers will opportunistically time the market and exploit mispricing by investing capital, issuing securities, or borrowing debt when costs of capital are low and shunning equity, divesting assets, repurchasing securities, and paying back debt when costs of capital are high. Rational managers will also incite mispricing, for example, cater to non-standard preferences of shareholders through earnings management or by transitioning their firms into an in-fashion category to boost the stock’s price.
The interaction of behavioral managers, that is, agents, with rational shareholders can also lead to distortions in corporate decision making. For example, managers may perceive fundamental values differently and systematically diverge from optimal decisions. Several personal traits, for example, overconfidence or narcissism, and environmental factors, for example, fatal natural disasters, shape behavioral managers’ preferences and beliefs, short or long term. These factors may bias the value perception by managers and thus lead to inferior decision making.
An extension of behavioral corporate finance is social corporate finance, where agents and principals do not make decisions in a vacuum but rather are embedded in a dynamic social environment. Since managers and shareholders take a social position within and across markets, social psychology and sociology can be useful to understand how social traits, states, and activities shape corporate decision making if an individual’s psychology is not directly observable.
Ingela Alger and Donald Cox
Which parent can be expected to be more altruistic toward their child, the mother or father? All else equal, can we expect older generation members to be more solicitous of younger family members or vice versa? Policy interventions often target recipients by demographic status: more money being put in the hands of mothers, say, or transfers of income from young to old via public pensions. Economics makes predictions about pecuniary incentives and behavior but tends to be agnostic about how, say, a post-menopausal grandmother might behave, just because she is a post-menopausal grandmother. Evolutionary theory fills this gap by analyzing how preferences of family members emerge from the Darwinian exigencies of “survive and reproduce.” Coin of the realm is so-called “inclusive fitness,” reproductive success of oneself plus that of relatives, weighted by closeness of the relationship. Appending basic biological traits onto considerations of inclusive fitness generates predictions about preferences of family members. A post-menopausal grandmother with a daughter just starting a family is predicted to care more about her daughter than the daughter cares about her, for example. Evolutionary theory predicts that mothers tend to be more altruistic toward children than fathers, and that close relatives would be inclined to provide more support to one another than distant relatives. An original case study is provided, which explains the puzzle of diverging marriage rates by education in terms of heterogeneity in preferences for commitment. Economists are justifiably loathe to invoke preferences to explain trends, since preference-based explanations can be concocted to explain just about anything. But the evolutionary approach does not permit just any invocation of preferences. The dictates of “survive and reproduce” sharply circumscribe the kinds of preference-related arguments that are admissible.
High-dimensional dynamic factor models have their origin in macroeconomics, more specifically in empirical research on business cycles. The central idea, going back to the work of Burns and Mitchell in the 1940s, is that the fluctuations of all the macro and sectoral variables in the economy are driven by a “reference cycle,” that is, a one-dimensional latent cause of variation. After a fairly long process of generalization and formalization, the literature settled at the beginning of the 2000s on a model in which (a) both n, the number of variables in the data set, and T, the number of observations for each variable, may be large; (b) all the variables in the data set depend dynamically on a fixed, independent of n, number of common shocks, plus variable-specific, usually called idiosyncratic, components. The structure of the model can be exemplified as follows:
where the observable variables are driven by the white noise , which is common to all the variables, the common shock, and by the idiosyncratic component . The common shock is orthogonal to the idiosyncratic components , the idiosyncratic components are mutually orthogonal (or weakly correlated). Last, the variations of the common shock affect the variable dynamically, that is, through the lag polynomial . Asymptotic results for high-dimensional factor models, consistency of estimators of the common shocks in particular, are obtained for both and tending to infinity.
The time-domain approach to these factor models is based on the transformation of dynamic equations into static representations. For example, equation () becomes
Instead of the dynamic equation () there is now a static equation, while instead of the white noise there are now two factors, also called static factors, which are dynamically linked:
This transformation into a static representation, whose general form is
is extremely convenient for estimation and forecasting of high-dimensional dynamic factor models. In particular, the factors and the loadings can be consistently estimated from the principal components of the observable variables .
Assumption allowing consistent estimation of the factors and loadings are discussed in detail. Moreover, it is argued that in general the vector of the factors is singular; that is, it is driven by a number of shocks smaller than its dimension. This fact has very important consequences. In particular, singularity implies that the fundamentalness problem, which is hard to solve in structural vector autoregressive (VAR) analysis of macroeconomic aggregates, disappears when the latter are studied as part of a high-dimensional dynamic factor model.
Nikolaus Robalino and Arthur Robson
Modern economic theory rests on the basic assumption that agents’ choices are guided by preferences. The question of where such preferences might have come from has traditionally been ignored or viewed agnostically. The biological approach to economic behavior addresses the issue of the origins of economic preferences explicitly. This approach assumes that economic preferences are shaped by the forces of natural selection. For example, an important theoretical insight delivered thus far by this approach is that individuals ought to be more risk averse to aggregate than to idiosyncratic risk. Additionally the approach has delivered an evolutionary basis for hedonic and adaptive utility and an evolutionary rationale for “theory of mind.” Related empirical work has studied the evolution of time preferences, loss aversion, and explored the deep evolutionary determinants of long-run economic development.
Ciaran N. Kohli-Lynch and Andrew H. Briggs
Cost-effectiveness analysis is conducted with the aim of maximizing population-level health outcomes given an exogenously determined budget constraint. Considerable health economic benefits can be achieved by reflecting heterogeneity in cost-effectiveness studies and implementing interventions based on this analysis. The following article describes forms of subgroup and heterogeneity in patient populations. It further discusses traditional decision rules employed in cost-effectiveness analysis and shows how these can be adapted to account for heterogeneity.
This article discusses the theoretical basis for reflecting heterogeneity in cost-effectiveness analysis and methodology that can be employed to conduct such analysis. Reflecting heterogeneity in cost-effectiveness analysis allows decision-makers to define limited use criteria for treatments with a fixed price. This ensures that only those patients who are cost-effective to treat receive an intervention. Moreover, when price is not fixed, reflecting heterogeneity in cost-effectiveness analysis allows decision-makers to signal demand for healthcare interventions and ensure that payers achieve welfare gains when investing in health.
Samuel Berlinski and Marcos Vera-Hernández
A set of policies is at the center of the agenda on early childhood development: parenting programs, childcare regulation and subsidies, cash and in-kind transfers, and parental leave policies. Incentives are embedded in these policies, and households react to them differently. They also have varying effects on child development, both in developed and developing countries. We have learned much about the impact of these policies in the past 20 years. We know that parenting programs can enhance child development, that centre based care might increase female labor force participation and child development, that parental leave policies beyond three months don’t cause improvement in children outcomes, and that the effects of transfers depend much on their design. In this review, we focus on the incentives embedded in these policies, and how they interact with the context and decision makers to understand the heterogeneity of effects and the mechanisms through which these policies work. We conclude by identifying areas of future research.
Alina Mungiu-Pippidi and Till Hartmann
Corruption and development are two mutually related concepts equally shifting in meaning across time. The predominant 21st-century view of government that regards corruption as inacceptable has its theoretical roots in ancient Western thought, as well as Eastern thought. This condemning view of corruption coexisted at all times with a more morally indifferent or neutral approach that found its expression most notably in development scholars of the 1960s and 1970s who viewed corruption as an enabler of development rather than an obstacle. Research on the nexus between corruption and development has identified mechanisms that enable corruption and offered theories of change, which have informed practical development policies. Interventions adopting a principal agent approach fit better the advanced economies, where corruption is an exception, rather than the emerging economies, where the opposite of corruption, the norm of ethical universalism, has yet to be built. In such contexts corruption is better approached from a collective action perspective. Reviewing cross-national data for the period 1996–2017, it becomes apparent that the control of corruption stagnated in most countries and only a few exceptions exist. For a lasting improvement of the control of corruption, societies need to reduce the resources for corruption while simultaneously increasing constraints. The evolution of a governance regime requires a multiple stakeholder endeavor reaching beyond the sphere of government involving the press, business, and a strong and activist civil society.
The geography of economic activity refers to the distribution of population, production, and consumption of goods and services in geographic space. The geography of growth and development refers to the local growth and decline of economic activity and the overall distribution of these local changes within and across countries. The pattern of growth in space can vary substantially across regions, countries, and industries. Ultimately, these patterns can help explain the role that spatial frictions (like transport and migration costs) can play in the overall development of the world economy.
The interaction of agglomeration and congestion forces determines the density of economic activity in particular locations. Agglomeration forces refer to forces that bring together agents and firms by conveying benefits from locating close to each other, or for locating in a particular area. Examples include local technology and institutions, natural resources and local amenities, infrastructure, as well as knowledge spillovers. Congestion forces refer to the disadvantages of locating close to each other. They include traffic, high land prices, as well as crime and other urban dis-amenities. The balance of these forces is mediated by the ability of individuals, firms, good and services, as well as ideas and technology, to move across space: namely, migration, relocation, transport, commuting and communication costs. These spatial frictions together with the varying strength of congestion and agglomeration forces determines the distribution of economic activity. Changes in these forces and frictions—some purposefully made by agents given the economic environment they face and some exogenous—determine the geography of growth and development.
The main evolution of the forces that influence the geography of growth and development have been changes in transport technology, the diffusion of general-purpose technologies, and the structural transformation of economies from agriculture, to manufacturing, to service-oriented economies. There are many challenges in modeling and quantifying these forces and their effects. Nevertheless, doing so is essential to evaluate the impact of a variety of phenomena, from climate change to the effects of globalization and advances in information technology.
The links of international reserves, exchange rates, and monetary policy can be understood through the lens of a modern incarnation of the “impossible trinity” (aka the “trilemma”), based on Mundell and Fleming’s hypothesis that a country may simultaneously choose any two, but not all, of the following three policy goals: monetary independence, exchange rate stability, and financial integration. The original economic trilemma was framed in the 1960s, during the Bretton Woods regime, as a binary choice of two out of the possible three policy goals. However, in the 1990s and 2000s, emerging markets and developing countries found that deeper financial integration comes with growing exposure to financial instability and the increased risk of “sudden stop” of capital inflows and capital flight crises. These crises have been characterized by exchange rate instability triggered by countries’ balance sheet exposure to external hard currency debt—exposures that have propagated banking instabilities and crises. Such events have frequently morphed into deep internal and external debt crises, ending with bailouts of systemic banks and powerful macro players. The resultant domestic debt overhang led to fiscal dominance and a reduction of the scope of monetary policy. With varying lags, these crises induced economic and political changes, in which a growing share of emerging markets and developing countries converged to “in-between” regimes in the trilemma middle range—that is, managed exchange rate flexibility, controlled financial integration, and limited but viable monetary autonomy. Emerging research has validated a modern version of the trilemma: that is, countries face a continuous trilemma trade-off in which a higher trilemma policy goal is “traded off” with a drop in the weighted average of the other two trilemma policy goals. The concerns associated with exposure to financial instability have been addressed by varying configurations of managing public buffers (international reserves, sovereign wealth funds), as well as growing application of macro-prudential measures aimed at inducing systemic players to internalize the impact of their balance sheet exposure on a country’s financial stability. Consequently, the original trilemma has morphed into a quadrilemma, wherein financial stability has been added to the trilemma’s original policy goals. Size does matter, and there is no way for smaller countries to insulate themselves fully from exposure to global cycles and shocks. Yet successful navigation of the open-economy quadrilemma helps in reducing the transmission of external shock to the domestic economy, as well as the costs of domestic shocks. These observations explain the relative resilience of emerging markets—especially in countries with more mature institutions—as they have been buffered by deeper precautionary management of reserves, and greater fiscal and monetary space.
We close the discussion noting that the global financial crisis, and the subsequent Eurozone crisis, have shown that no country is immune from exposure to financial instability and from the modern quadrilemma. However, countries with mature institutions, deeper fiscal capabilities, and more fiscal space may substitute the reliance on costly precautionary buffers with bilateral swap lines coordinated among their central banks. While the benefits of such arrangements are clear, they may hinge on the presence and credibility of their fiscal backstop mechanisms, and on curbing the resultant moral hazard. Time will test this credibility, and the degree to which risk-pooling arrangements can be extended to cover the growing share of emerging markets and developing countries.
Noémi Kreif and Karla DiazOrdaz
While machine learning (ML) methods have received a lot of attention in recent years, these methods are primarily for prediction. Empirical researchers conducting policy evaluations are, on the other hand, preoccupied with causal problems, trying to answer counterfactual questions: what would have happened in the absence of a policy? Because these counterfactuals can never be directly observed (described as the “fundamental problem of causal inference”) prediction tools from the ML literature cannot be readily used for causal inference. In the last decade, major innovations have taken place incorporating supervised ML tools into estimators for causal parameters such as the average treatment effect (ATE). This holds the promise of attenuating model misspecification issues, and increasing of transparency in model selection. One particularly mature strand of the literature include approaches that incorporate supervised ML approaches in the estimation of the ATE of a binary treatment, under the unconfoundedness and positivity assumptions (also known as exchangeability and overlap assumptions).
This article begins by reviewing popular supervised machine learning algorithms, including trees-based methods and the lasso, as well as ensembles, with a focus on the Super Learner. Then, some specific uses of machine learning for treatment effect estimation are introduced and illustrated, namely (1) to create balance among treated and control groups, (2) to estimate so-called nuisance models (e.g., the propensity score, or conditional expectations of the outcome) in semi-parametric estimators that target causal parameters (e.g., targeted maximum likelihood estimation or the double ML estimator), and (3) the use of machine learning for variable selection in situations with a high number of covariates.
Since there is no universal best estimator, whether parametric or data-adaptive, it is best practice to incorporate a semi-automated approach than can select the models best supported by the observed data, thus attenuating the reliance on subjective choices.