Please enable JavaScript.

Coggle requires JavaScript to display documents.

The New Economic History (Economic History and Cliometrics (Louis P. Cain…

- - - - the National Bureau of Economic Research
        
        To begin bridging this gap, the National Bureau of Economic Research (NBER) was founded in 1920 to produce research firmly based on economic facts – quantitative if possible, scientific and impartial, and neutral with respect to ideology and policy. During the interwar period the NBER added substantially to the statistical infrastructure for examining american economic development. Among the most important studies was the series on income, savings and expenditure in the United States by Simon Kuznets. Kuznets was one of the three economic historians to win the nobel price in economics along with Fogel and Douglass North.
      - finding data for economic research
        
        In the first half of the twentieth century, economic historians rarely appealed to explicit economic theory, but relied more on an accumulation of facts and anecdotes in describing and
        attempting to explain the past. Economists on the other hand were often accused of relying too much on theory and not enough on facts. Testing economic theories using historical data was difficult as data was scarce. Important economic statistics, including series on prices and wages, production and employment (and thus unemployment) were few and far between, partly because the cost of retrieving them were prohibitively high in that pre computer age. Essentially: data for economic theory was very expensive
      - Research on economic growth after WWII
        
        After the Second World War, economists increasingly focused their research on economic growth and development in hopes of preventing a return to Depression-era conditions in developed countries and of improving the lot of less developed countries. Because understanding growth requires understanding history, a new generation of scholars trained in economics departments turned their attention to economic history. When they began to read the works of older economic historians, many complained that important economic issues had been misinterpreted because of a failure to conjoin economic theory and historical data, and they pushed their colleagues to employ economic theory along with measurement.
    - - the economics of slavery in the ante-bellum South
        
        The value of the new approach was convincingly displayed in “The economics of slavery in the ante-bellum South”, one of the most influential and seminal works of cliometrics, which was then called “the new economic history.” Rather than examine the mindset and motives of slaveholders, authors Alfred H. Conrad and John R. Meyer (1958) approached the subject as an investment problem. Was slavery profitable? This question had been debated without resolution for over a century. Conrad and Meyer modeled slave owners as investors with a capital asset subject to depreciation and maintenance costs. They examined historical data on cotton yields, slaves’ average life spans, and prices of inputs and outputs – and calculated that owning slaves was more profitable than many other potential investments. Their finding (and extensions of it) helped create a new, lasting consensus concerning the profitability of slavery.
      - Douglass North noting 6 years later
        
        A revolution is taking place in economic history in the United States. It is being initiated by a new generation of economic historians who are both skeptical of tradi-tional interpretations of U.S. economic history and convinced that a new economic history must be firmly grounded in sound statistical data. Even a cursory examination of accepted “truths” of U.S. economic history suggests that many of them are incon-sistent with elementary economic analysis and have never been subjected to – and would not survive – testing with statistical data.
      - works of the revolution
        
        North could point to Conrad and Meyer’s work on the profitability of slavery, Robert Fogel’s (1964) work finding that railroads were not indispensable to nineteenth-century American economic growth, research on the unimportance of the Civil War for accelerating U.S. indus-trialization (Gallman 1960), and several other projects to support the existence of a revolution.
      - resolved many things that had previously been unclear
        
        In those years, the cliometric approach quickly resolved several ongoing debates in American economic history. For example, using a simple trade model, estimates of demand and supply elasticities, and colonial prices, the per capita burden of Britain’s Navigation Acts as a per cent of colonial income was estimated to be much smaller than many supposed. Hence it was an unlikely spark for the American Revolution (Thomas 1965; McClelland 1969). Likewise, Peter Temin’s (1968) analysis of sources of changes in the money supply demonstrated that the inflation and financial crisis of the 1830s had their origin in international events largely beyond President Andrew Jackson’s control. Those crises would have taken place whether or not he had acted as he did regarding the Second Bank of the United States – thus fundamentally reinterpreting another important episode in American economic history.
    - - The cliometric approach aimed to replace imprecise qualitative statements – such as “It is diﬃcult to exaggerate the importance of this” – with precise and accurate estimates of economic magnitudes and economic relationships. As cliometricians harnessed computers, which allowed them to quickly calculate totals and averages, they also tried to estimate underlying economic relationships and marginal eﬀects using voluminous (but previously silent and unu-sable) bodies of archival data. As some debates were settled, however, the new approach opened even more research questions to analysis and scrutiny.
      - popularity of cliometrics
        
        One way to gauge the progress of cliometrics is to examine the pages of the leading eco-nomic history journal. In the early 1950s, less than 2 per cent of the pages in the Journal of Economic History were in cliometric articles – those which used measurement (seen in a growing number of tables) and explicit economic theory. This figure subsequently climbed to 10 per cent in the late 1950s, 18 per cent in the early 1960s, 44 per cent in the late 1960s, and 78 per cent in the early 1970s (Whaples 1991).
        
        In the late 1950s, cliometrics was seen by some as a passing fad, but by the 1970s it was the standard operating procedure for American economic historians. In the following decades, cliometrics was increasingly adopted by scholars from out-side the U.S. as well. Many of the achievements of this first generation of cliometric research were surveyed in Fogel and Engerman (1971), McCloskey (1976), Lee and Passell (1979), and more recently (and retrospectively) Lyons et al. (2008) – all of which (sometimes breathlessly) emphasized both the novelty and important breakthroughs from the approach as it gained dominance within the United States.
  - - - The cliometric approach has had a natural appeal to historically minded researchers trained in departments of economics. In one sense, it is simply the application of standard economic tools – initially developed to understand the present – to the past. However, cliometrics has been more controversial among scholars trained in history departments.
    - - One of the strongest early criticisms of the new method came from Fritz Redlich. In pub-lished work he expanded on a point he made most succinctly in responding to a question at an academic conference. The young questioner asked the old professor, “What is your model?” Redlich responded, “Model? I deal with the truth.” One of Redlich’s most telling arguments was that cliometricians focused too much on purely economic considerations and ignored the impact of institutions.
      - He was even more strident in his attack on the use of explicit counterfactuals, which played a prominent role in the early history of cliometrics, especially in the case of Robert Fogel’s influential analysis of the impact of railroads. To fully assess this impact, Fogel modeled a counterfactual world in which railroads never existed. Among other things, this required him to project the expansion of navigable waterways that would have been built had there been no railroads, and to project the consequences of that expansion, such as shifts in population. Redlich attacked such counterfactuals by calling them “figments.” “Hypotheses are based on assump-tions which are held to have a counterpart in reality, while figments are assumptions having no such counterparts or at least known to be irrealistic” (Redlich 1965: 484). Due to their basis in reality, hypotheses can be verified or falsified; figments can only be justified. Figments lead to economic models, and “a model is never a piece of history” (Redlich 1965: 490).
      - response of the proponents
        
        Proponents of the cliometric approach rebutted this criticism by arguing that every historian has a model. Without mental models to simplify the bewildering array of sensory facts, a person wouldn’t be able to make it through the day. Without models to simplify the bewildering array of historical facts, no scholar could make sense of the past. Cliometricians, such as Lance Davis (1966), argued that their breakthrough was to make explicit the models that had once been implicit. For example, those who argue for the indispensability of railroads in nineteenth-century American economic growth implicitly have a model that suggests the demand for transportation was massive, that railroads oﬀered immense advantages over other modes of transportation, andthat no suitable substitute for railroads was likely to have performed as well. Fogel (1964) sub-jected this model to empirical tests and found it wanting. He then formalized an alternative set of assumptions about transportation demand and the shift in supply brought about by the rail-roads. Cliometricians celebrated this conceptual breakthrough, but debated whether or not the theoretical assumptions in Fogel’s model accurately reflected the structure of the economy and whether the numbers used in it were accurate. The conclusion of the profession – especially those trained in economics – was that Fogel’s modeling and measurement were convincing.In a 1995 survey, Whaples (1995: 143) found that 89 per cent of economic historians in eco-nomics departments rejected the idea that “without the building of the railroads, the American economy would have grown very little during the nineteenth century.” Economic historians housed in history departments (a dwindling breed) were less convinced: 66 per cent of them rejected the idea.
    - - Redlich was not alone in his criticism. Cliometricians were mostly economists, and their papers had to conform to economists’ standards. Their method and style, and assertions of the virtues of cliometrics, alienated older economic historians in particular, and historians in general. American Historical Association President Carl Bridenbaugh lamented that some colleagues had succumbed “to the dehumanizing methods of social sciences,” and warned against worshipping “at the shrine of that Bitch-goddess, QUANTIFICATION” (Bridenbaugh 1963: 326). To many contemporary American economic historians, these arguments are a quaint episode. Cliometrics is their orthodoxy; as they see it, cliometricians “won the West,” imposing a “Pax Cliometrica” (McCloskey 1987: 77).
      - artefact of modern, western thinking
        
        Boldizzoni
        
        A more lasting criticism has been that the default model of cliometrics is an artifact of modern, Western thinking. Boldizzoni (2011) argues that cliometrics hasn’t improved our under-standing of the past but has distorted it, because its default assumption is that there are essentially universal laws of human behavior that have held at all times and in all places. Economists, he argues, have projected their belief that people are “rational maximizers” from the modern West – where the assumption may fit fairly well – to other times and places when and where it doesn’t fit.
        
        North has complained that cliometrics needs to focus more on institutions and their evolution. In a series of works designed to get to the fundamental forces behind long-term economic trends – including both pre-modern economic stagnation and modern growth – North and his co-authors (North 1981, 1990, 2005; North et al. 2009) pushed their analysis further and further back, but they never abandoned their key assumption that individuals acted out of self-interest and were constantly cognizant of costs versus benefits in making decisions.
        
        At heart, Boldizzoni attacks cliometrics for relying on ill-considered deductions rather than wading deeply into the details of historical life and using induction to understand the logic of earlier societies and their economies. Cliometricians would rebut this by arguing that their method relies on both deduction and induction. Accordingly, the next section of this chapter looks at the practice of modern cliometrics and the debates among cliometricians about how to make conclusions regarding historical cause and eﬀect.
  - - - Perhaps the most common practice in cliometrics is to use multivariate regression analysis, which produces an estimate of the marginal impact of a variable of historical interest while holding other important factors constant. A typical example is Buﬀum and Whaples (1995) who seek to test the magnitude of employee-based discrimination in the labor market of the late 1800s using data on furniture workers collected by the Michigan Bureau of Labor Statistics. Their study begins with a simple economic model that assumes this market was competitive so that wages reflected workers’ productivity, but that workers were also compensated for any negative workplace conditions. Using wages as the dependent variable and holding constant measurable worker characteristics (such as occupation, education, and experience) likely to influence productivity (and therefore wages), this study examines the marginal impact of variation in the ethnicity of co-workers. It finds that a one percentage point increase in the share of the work force who were not from a worker’s own ethnic group increased his wage about 0.1 per cent. This is interpreted as evidence that employers had to pay their employees more when they were in close contact with co-workers they liked less (or disliked more) than members of their own group. The study also finds that workers from Protestant ethnic groups were generally paid more when working with Catholics, but concludes that the additional labor costs generated by this employee-based discrimination were probably oﬀset by several benefits that rendered complete segregation unnecessary.
      - This study is typical of cliometric work that applies a widely used economic model to his-torical data. The regression coeﬃcients allow the economic historian to go an important step beyond contemporary and historical accounts that agreed workers in these groups often didn’t get along and disliked each other. Regression coeﬃcients quantify how much workers disliked each other by estimating how much more they had to be paid to work with one another – or how much pay they were willing to sacrifice to work with members of their own group. The study reaches the cliometric goal by illuminating the past and also breaking ground in testing a theoretical implication – that if employees discriminate against each other they must be paid more to work with each other – that can’t be tested well using modern data but can be tested with a rich historical data set.
    - - problems
        
        A more diﬃcult task is to examine a historical change, an episode or a public policy, and convincingly determine its causes or consequences. Data sets on workers’ wages often include thousands of observations, and it is clear that some explanatory variables are exogenous (that is, caused by factors outside the system). Wage regressions can be fairly straightforward to interpret because it is clear that some variables cannot influence others, making conclusions about cau-sation more convincing. When wages are higher for middle-aged workers than for younger workers, it’s fairly clear that age is driving wage and not the other way around – that getting older pushes up one’s wage since earning more can’t cause one to become older. However, examining the causes and consequences of events and policies is harder because it’s more diﬃcult to find plausibly exogenous determinants, and because events can be so complex that an unmeasured, omitted variable might be at work.
      - panel data as a solution
        
        Panel data can help solve this problem. A panel generally consists of two dimensions – a spatial dimension that consists of cross-sectional observations (e.g. nations, states, cities, indus-tries, firms, groups of people, or individuals) and a temporal dimension that consists of time-series observations (periodic measures of characteristics of what is being observed).
        
        example
        
        Consider this question: Did the riots in American cities in the 1960s cause long-term eco-nomic harm to those cities? Conventional economic theory suggests this is plausible, because a property’s value is believed to reflect the discounted value of the expected net flow of utility associated with its ownership – including not only the physical quality of the structure, but also such things as security, proximity to work, family, friends and shopping, the quality of muni-cipal services, and the taxes needed to support such services. In theory, if a riot causes a sus-tained decline in the perceived amenities in one location relative to others, it would register as a relative decline in property values. In line with this theory, examining data shows that – holding constant other factors like region, population, crime rate, segregation, and the trend in property values – property values fell more in cities hit by more severe riots (Collins and Margo 2007).
        
        But what if the riots themselves were more severe in cities that were economically weaker, and this economic weakness accounts for both the lagging property values and the riots? In cases like this, economic historians look for “natural experiments” – changes or diﬀerences that plausibly arise outside the economic system, which allow one to compare situations or places that received one “treatment” with those that received another “treatment.” Collins and Margo report contemporary observations that rainy weather made rioting less severe, and then estimate the impact of city-specific weather in the period following Martin Luther King’s assassination – showing that (holding other factors constant) more rain in April 1968 led to less rioting. In this case, rain is used as an “instrumental variable,” an exogenous variable that does not itself belong in the explanatory equation but which is correlated with the endogenous explanatory variables. Putting these pieces together (using a two-stage regression estimation framework), Collins and Margo convincingly confirm that more severe rioting had an independent, large, negative impact on urban property values. They estimate that riots decreased black-owned home values in these cities by about 10 per cent.
        
        most famous panel data
        
        Among the panel data sets that are widely used in economic history are the Panel Study of Income Dynamics (PSID), the National Longitudinal Survey of Youth (NLSY), and the Union Army data of the Center for Population Economics. The PSID is a panel of U.S. families that began in 1968 and continues today. Economic, social, and health factors have been measured over individuals’ lives and across generations. The NLSY is one of several panels collected by the Bureau of Labor Statistics. The most widely used of these is the National Longitudinal Survey of Youth 1979, which surveys young men and women born in the years 1957–64. The Union Army Data Set (and the U.S. Colored Troops Data Set) consists of white (and black) males mustered into the Union Army during the Civil War. Military, socio-economic, and medical information from several sources throughout their lifetimes have been collected and are part of ongoing research eﬀorts.
        
        construction of own panels
        
        Many economic historians have constructed their own panels. An important example is the panel of city-, county- and state-level New Deal spending, economic activity, and demographic data constructed by Price Fishback and his co-authors. For example, Fishback et al. (2007) examine the impact of New Deal spending on infant mortality, non-infant mortality, and general fertility rates in major U.S. cities between 1929 and 1940. They include controls for city characteristics, city and year fixed eﬀects, and make use of instrumental variables (voting patterns in years prior to the Great Depression, and Congressional committee assignments, for example) that helped determine the allocation of New Deal spending. They conclude that about $2 million (measured in year 2000 dollars) in additional relief spending was associated with a reduction of one infant death, half a homicide, one suicide, 2.4 deaths from infectious disease, and one death from diarrhea in each large urban area.
        Finding a convincing instrumental variable can be very challenging. The researcher needs to find a factor that helped drive the economic events but did not arise simultaneously from within the economic system. When it comes to policy analysis, this can be almost impossible because political decisions are so intricately intertwined with economic forces.
  - - - The typical regression only says something about correlation, but Granger won the Nobel Prize in Economics for arguing that, under certain conditions, the regression can also say something about causation. In brief, a time series of one variable is said to Granger-cause a time series of another variable if it can be shown that the first variable’s values provide statistically significant information about the future values of the second variable – thus, if one time series can be used to forecast another, then one can argue that it helps cause it to vary, at least in a statistical sense. The first mention of such a test in the economic history literature appears in Michael Edelstein’s (1980) discussion of a paper by Michael Bordo and Anna Schwartz at the annual Economic History Association meetings. Edelstein suggests the use of the test, but in the published paper Bordo and Schwartz (1981) argue that, while the Granger test could be used to test causality between money and the price level, or the price level and the terms of trade, they believe there is an omitted common influence aﬀecting both the terms of trade and the price level, and this means the test is inappropriate. However, these tests are a main focus of Schwartz (1981) and her contribution toward revisiting the Great Depression contained in Brunner (1981). Her long-run results strongly suggest money Granger-caused income and not the reverse: the supply of money was not endogenously driven by the business cycle.
      - stationary time series
        
        Stationarity is an attribute of a time series in which the mean and standard deviation are constant over time. It is important since standard econometric tools assume finite variances for the variables of the estimated model. Stationarity provides this finite variance. Thus, the first step in analyzing time-series data is to test if one’s data are stationary or not. Its first appearance among economic historians is David Pope’s (1984) analysis of Kondratieﬀ cycles in Australia. The most common way to test for (non-)stationarity is to apply a Dickey-Fuller test that checks whether a unit root is present in an autoregressive model. In his examination of open market operations in the 1920s, Mark Toma (1989: 116) used this test to argue that, in a regression of the level of Federal Reserve security holdings against its lagged value and a constant, “standard test procedures reject the non-stationarity hypothesis that the lagged coeﬃcient equals one.” Within two years, such tests were common in the economic history journals.
      - cointegrated time series
        
        Two (or more) time series are said to be cointegrated if they move together, if they statisti-cally share a common stochastic drift. Alternatively, two time series are cointegrated if they are non-stationary but a linear combination of the two series is stationary. Gary Libecap (1989) examines interstate cartel production coordination in the petroleum market. If residual pro-duction by Texas was for the purpose of maintaining nominal crude oil prices, there would have to be a long-term relationship between output in Texas and market demand in the U.S. – that is, the two time series would be cointegrated. Libecap finds no statistically significant evidence that this was true.
    - - One of the major issues confronted by time-series analysis has been to identify the timing and potential causes of the British Industrial Revolution. Many scholars have contributed to this literature, and among the major contributors are David Greasley and Les Oxley (2010). Extending the “best guess” estimates of British Industrial Production generated by Nick Crafts and Knick Harley (1992) and using an augmented Dickey-Fuller test, Greasley and Oxley (2010) divide almost 300 years of British history into five epochs: 1700–80; 1781–1851; 1852– 1913; 1922–73 and 1974–92. Their times-series tests characterize the period between 1700 and 1913 as alternating between trend stationary, diﬀerence stationary, and trend stationary, so they make a case for dating the British Industrial Revolution to the second epoch, 1781–1851. As to the causes of British industrialization, their tests indicate that cotton and iron were the leading industries, and coal was a follower. Cotton had wider linkages than iron, “statistically causing paper and shipbuilding production,” but they do not regard it as more important in the 1815–60 period “as bidirectional causality between iron goods and cotton output lies at the heart of the cotton, mining and metals sector” (Greasley and Oxley 2010: 1009–10). (This paper also contains an extensive overview of time-series methods used by economic historians.)
      - techniques in general
        
        It can be argued that many of these techniques help us to understand the statistical process that generated the data, but that the statistical process is not necessarily the economic process and Granger-causality is not economic causality. Macroeconomists have generated competing models of the same economic process: there is no single “correct” understanding. It is safe to say that there is much work still to be done to understand and dynamically model the economy.
        These examples are only a small sample of the empirical inquiries to which time-series econometrics has been employed by economic historians. Given the importance of time series to economic history and the time-series data economic historians have produced, it seems inevi-table that new statistical techniques for analyzing time series, as they become available, will be adopted quickly.
  - - - As historical questions have become more complicated, the models used to answer them have become more complex in accounting for feedback mechanisms and impacts that spread throughout the economy over time. For example, the economic eﬀects of immigration ripple throughout an economy, so a comprehensive model that can capture this complexity is needed. A naïve, theory-free approach may suggest that one can isolate the impact of historical immi-gration on wages by simply comparing wages in cities and states where immigrants have settled with those in places they haven’t. Unfortunately, this conveys virtually nothing for two reasons. First, immigrants don’t usually choose their destinations randomly; given their mobility, they often select areas that have the best job-market opportunities. They are attracted to the highest-wage cities and states, so finding that wages are higher in places with a lot of immigrants could confuse cause and eﬀect. Second, suppose that immigrants flood into a city – increasing its labor supply relative to its labor demand – pushing wages in the city down. The native-born population is not likely to react passively to this, but will begin to leave this market, entering other markets and pushing down wages in places where immigrants haven’t gone. If these native-born workers are mobile enough, just enough of them will leave to make the wages in all cities approximately equal for similar jobs. The impact of immigration will be diﬀused across the entire economy, so looking for its impact by comparing where immigrants are and aren’t located can be erroneous.
    - - To capture the impact of immigration and other broad economic changes like trade and technological change, economic historians have turned to models of the economy that can capture the feedbacks. One influential immigration study is Hatton and Williamson (1998), which includes a computable general equilibrium model (developed in O’Rourke et al. 1994) of the interactions between the U.S. and UK economies during the late 1800s and early 1900s. In the U.S. model, output and factor prices are endogenously determined – that is, they arise within the system. Consumers are constrained by endowments and maximize a utility function; their income and expenditures are endogenous. There are four production activities that obey standard production functions – food, agricultural intermediates, manufacturing, and services; three primary factors of production – land, labor (divided into agricultural and non-agricultural),
      - and capital; two imported goods; and foreign exchange. The model assumes that firms minimize costs (to maximize profits); competition is perfect; capital is perfectly mobile across sectors; non-agricultural labor is perfectly mobile between manufacturing and services; and rural–urban migration is determined endogenously by wage gaps. The model also includes an explicit con-sumer utility function. Equilibrium occurs when for every sector price equals costs; for every commodity, demand equals supply; and the consumers’ incomes equal the rents on all endow-ments. The thirteen prices are endogenously determined, as are the nine activity levels. The UK model is similar but somewhat simpler. The model allows its users to examine the impact that labor flows from Britain to the U.S. had throughout the economy – on wages and other factor price levels, on prices of goods, on the allocation of resources, and more.
      - criticisms of the model
        
        While a model such as Hatton and Williamson’s can be an advance over a simpler partial-equilibrium model, it is open to many criticisms. Rosenbloom (1998) finds Hatton and Wil-liamson’s conclusion that trans-Atlantic migration (along with trade) was an important factor in explaining the narrowing of international wage gaps during this period is plausible; yet he cautions that it is diﬃcult to adequately assess these conclusions in part because of the very complexity of the general equilibrium models on which they rest. Are the key assumptions of the model – the elasticities, interactions, and functional forms embedded in it, the calibrations of key parameters – a good reflection of reality? The very complexity of the economy necessitates the use of such complicated models, but also undermines their usefulness – and these complex models are simplifications that leave out much that may be important (e.g. immigration’s impact on technological progress, culture, and institutions). Some important events in economic his-tory – like immigration – have such complex and far-reaching impacts that even the most advanced cliometric techniques can’t adequately assess them. But the bottom line is that models by their very design and nature are abstractions from the complexity of life. The job of the economic historian is to push refutable hypotheses as far as the theory and data will allow, and to confirm or reject these hypotheses based on the credible evidence that can be marshaled. Perhaps there will never be the tools to adequately address the most important and complex historical economic questions. Researchers probably will not innovate enough to create ade-quate methods to completely address these important questions. But it is essential to press the available technology as hard and as fully as we are able.
- - - - Economic history is a subset of history. Both economists and historians are trying to tell plausible stories about the past, and they succeed or fail by narrative standards to connect one event to another. Cliometricshas transformed the study of economic history from a narrative to a mathematical format.
    - - In the process, cliometricians have contributed to the development of both economics and history by combining theory with quantitative methods, constructing and revising databases, and adding the variable of time to traditional economic theories. This has made it possible to question and reassess earlier findings, thus expanding the frontier of our knowledge of the past and its ability to portend the future.
    - - The use of history as a crucible to examine economic theory has deepened our knowledge of how, why and when economic growth and development has and will occur.
    - - The rise of cliometrics, originally known as ―new‖ economic history, has fostered much debate about the proper balance of economics and history in economic history.
      - Gavin Wright
        
        In 1971 Gavin Wright surveyed the development of cliometrics. He argued that the ―new‖ economic history was more about the use of economic theory than the use of econometrics in the study of history.1 In agreement,
      - Difference between the new and old Economic History
        
        Deirde McCloskey
        
        Deirdre McCloskey defined a cliometricianas an economist applying economic theory to history. She claimed that it was not the sophistication of the model, but the use of a model at all that differentiated the ―new‖ economic history from the old.2
        
        Thomas Cochrane
        
        Thomas Cochrane portrayed the difference between the traditional practice of economic history and the new cliometric movement as a ―controversy over types of models: the old say that realistic models usually have to be too highly generalized or too complex to allow the assumption of mathematical relationships; the new are primarily interested in applying operative models to economic data
  - - - Clio‘s roots are historical in nature, and its focus on theory has actually come full circle over the last century and a half. A mathematical movement in the economics discipline, advanced computing technology, and a shift in the focus of the role of history within economics all contributed to the proliferation of the ―new‖ economic history that rewrote the landscape of the discipline. The emphasis on theory and formal modeling that distinguishes cliometrics from the ―old‖ economic history now blurs the distinction between economic history and economic theory.
    - - Cliometrics today is closely related to, but not necessarily the same thing as its progenitor, economic history. While there is considerable overlap between the membership of the Cliometric Society and its American brethren, the Economic History Association (EHA), the latter has many more members who reside in history departments than does the Cliometric Society. Indeed, one of the great criticisms of the cliometric movement is the wedge that it has driven between the practitioners of economic history in history and economics departments due to its focus on quantitative measures and neoclassical theory.4
    - - In his inaugural presidential address to the EHA, Edwin Gay noted that economic historians needed to wed the skills of the economist with the historian in order to accomplish their task. He believed the molding of these two skillsets were essential, but not easy to accomplish.6 That has notchanged over the past three quarters of a century. What has changed is the degree to which those economic skills have become more formalized and technically demanding.
    - - The clash between cliometricians and historians today is not all that different from the clash between economists and historians that began in the 19th century. Carl Menger compared historians to foreign conquerors, complaining that they were forcing their terminology and methods on economists.7 Half a century later, Thomas Ashton accused those who objected to the idea that economic theory should be applied to history of not truly understanding the nature of economics. (Menger)
    - - Economic history originated largely as a revolt against classical theory and in its early years it shunned the use of statistical techniques. By the 1920s the attitude toward theory and statistics began to soften. Cliometrics is the continuation of this theoretical-quantitative tradition, now fortified by advances in economic theory, the melding of economics with approaches from other disciplines, and the growth of computing power. The latter has had profound impacts on the ability to analyze and disseminate data.
  - - - To date, the main achievements of cliometrics have beento slowly but surely establish a solid set of economic analyses of historical evolution by means of measurement and theory, and, following the path blazed by Douglass North, to recognize the limits of neoclassical theory and bring into economic models the important role of institutions. Indeed, this latter focus ultimately spawned a new branch of economics altogether, the new institutional economics. Nothing can now replace rigorous statistical and econometric analysis based on systematically ordered data. Impressionistic judgements supported by doubtful figures and inadequate methods padded by subjective impressions have now lost all credibility.
    - - Indeed, the most vigorous new trend in the social sciences is without a doubt the preoccupation with quantitative and theoretical aspects. It is the feature that best distinguishes the concepts of the current generation of scholars from its forbears. Even the most literary of our colleagues is ready to agree to this. There is nothing surprising about this interest. One of the characteristic features of today‘s younger generation of scholars is most certainly that their intellectual training is much more deeply marked by science and the scientific spirit than that of the generations that preceded them. It is, therefore, not surprising that young scientists should have lost patience with regard to the tentative approach of traditional historiography and have sought to build their work on foundations that are less ―artisanal.‖
      - Human and social sciences are thus becoming much more elaborate in the technical respect, and it is difficult to believe that a reversal of the trend is likely to occur. However, it is also clear that a significant proportion of human and social scientists have not yet accepted the new trends aimed at using more elaborate methodology and clear concepts conforming to new norms in order to develop, in the tradition of Robert Fogel, a truly scientific human and social science.
    - - Economic history has enjoyed a resurgence due to a number of recent topics, including financial crises, the increasing divergence in living standards, new institutional economic history, anthropometric history, and environmental history.The contributions of cliometrics can be broadly categorized as either technique, data, revision, or new approaches.
    - - technique
        
        Technique is what many people think of when they hear the term cliometrics. Certainly the advancement of econometric theory and computing power have contributed greatly to the techniques used by cliometricians. In the early issues of the Journal of Economic History the appearance of an equation was rare. As the ―new‖ economic history took hold under the editorship of Douglass North and William Parker in the early 1960s, equations and the occasional OLS regression began to appear regularly. Today, that occasional OLS regression has been replaced by the latest econometric advancement. It is now a rare article in the JEH that does not rely on the latest econometric advances as part of its analysis. However, technique is not merely mathematical sophistication at its utmost. One of the earliest and still useful techniques available to the cliometrician is the counterfactual.
        
        counterfactual
        
        One of the earliest and still useful techniques available to the cliometrician is the counterfactual.
        
        definition
        
        Counterfactual analysis is the idea of determining the impact of an event or factor by considering what would have happened in its absence.
        
        Fogel
        
        Fogel was not the first to use this form of identifying opportunity costs,10 but he was the most extensive user of it and became famous for his use of the technique in his landmark railroad study. He attempted to estimate how much less developed the America economy would have been had there been no railroads. Although historians were familiar with counterfactual arguments, the idea of an explicit counterfactual measurement was still a foreign notion in the early 1960s. The concept of social savings is itself an important research tool, but Fogel greatly advanced its significance by defining it operationally, so that his calculations could be tested against alternative estimates and possible alternative definitions.The publication of Fogel's railroad study generated an entire subdiscipline of parallel studies and, more importantly, provided a methodological foundation for the systematic study of economic history and long-term economic growth.
        
        Fogelshowed how well economic history could benefit from the careful application of theory and econometrics. The work immediately generated substantial controversy, and even today some quibbling over minor details occurs. However, time has failed to overturn Fogel‘s major conclusions: that per capita income growth would have been set back only a few months had the railroads never been invented, and that no other industry was likely to have been more important than the railroads. Since its publication, the great majority of economic history has been written by scholars employing those basic economic and econometric tools.
        
        definition of the methodological features
        
        Fogel defined the methodological features of cliometrics. He considered it fundamental that cliometrics should stress measurement while recognizing the existence of close links between measurement and theory. Indeed, unless it is accompanied by statistical and/or econometric processing and systematic quantitative analysis, measurement is just another form of narrative history. It is true that it replaces words with figures, but it does not bring in any new factors. In contrast, cliometrics is innovative when it is used to attempt to model all the explanations of past economic development. In other words, the main characteristic of cliometrics is the use of hypothetico-deductive models that call on the closest econometric techniques with the aim of establishing the interaction between variables in a given situation in mathematical form.
        
        constructing a model
        
        This generally consists of constructing a model—of general or partial equilibrium—that represents the various components of the economic evolution in question and showing the way in which they interact. Correlations and/or causalities can thus be established to measure the relative importance of each over a given period of time.
      - data
        
        big data
        
        Cliometrics has also contributed large and expansive data sets for researchers. The accumulation of the data is in itself monumental in many respects, but its usefulness has been expanded by the rapid growth of computing power. The ability to handle ―big data‖ is not a cliometric issue by itself, but the construction of significant, important historical data sets, which can then be analyzed using the latest econometric techniques and computer programs, is very much a contribution of cliometrics
        
        lack of data?
        
        Fogellong argued that the lack of relevant data more than the lack of relevant theory is often the greater hurdle in historical research. In this way, economic historians have made some of the greatest contributions to the fields of economics and history by discovering and compiling new data sets that can then be used by future researchers to better understand the evolution and growth of economies over time.
        
        Fogel unifying history and economics
        
        Fogel pioneered the use of large-scale cross-sectional and longitudinal data sets culled from original sources to examine current policy issues, and his multitude of graduate students went on to do the same in their careers.12Robert Fogel reunified economics and history. ―Using the best techniques of modern economics and gathering the widest samples of historical data he has reinterpreted American economic growth. Rather than conjecturing on the causes of growth he has asked persistently ―how large?‖ and seen the way to answer. He has set a new standard for empirical thoroughness in economics and anew standard of logical cogency in history. The quantitative history he advocates has opened new ways to the past. The historical economics he helped create, an economics made wiser by a knowledge of history, brought economics back to the larger questions
        
        Cliometrics has also contributed large and expansive data sets for researchers. The accumulation of the data is in itself monumental in many respects, but its usefulness has been expanded by the rapid growth of computing power. The ability to handle ―big data‖ is not a cliometric issue by itself, but the construction of significant, important historical data sets, which can then be analyzed using the latest econometric techniques and computer programs, is very much a contribution of cliometrics
        
        Edwin Gay
        
        But Fogel was not the first to recognize the importance of reliable, comprehensive data sets. During his service to the U.S. government during WWI, Edwin Gay became convinced ofthe need for better economic statistics. He and Wesley Mitchell headed the Central Bureau of Planning and Statistics, responsible for the gathering and reporting of statistical data. Together they helped found the NBER to stimulate the collection and interpretation of historical statistics.
        
        The NBER ultimately served as a catalyst for the change in emphasis from narrative to quantitative studies in economic history. Mitchell, Simon Kuznets, Arthur Burns, Solomon Fabricant, and Harold Barger produced a series of quantitative descriptions of American economic growth while at the NBER that measured growth as far back as the 1870s. The culmination of this quantitative approach to descriptive economic history was theHistorical Statistics of the United Statesproduced by a committee of scholars andpublished by the U.S. Census Bureau in 1960.
        
        Mitchell served as research director at the NBER for the first quarter century of its existence.14He gathered tremendous amounts of empirical economic data in order to draw inductive generalizations from it, and issued an urgent call for more data collection from around the world. The NBER was central to this data collection effort and served as a sort of haven for statistical economists. The mission of the NBER was to gather empirical information of many kinds about the American economy in order to create a robust foundation for theoretical generalizations.
      - revisionism
        
        A clear understanding of the causes of economic growth is among the most important things an economic historian can do. Cliometrics has overturned some accepted wisdoms and in the process caused hard feelings, resentment, and controversy. However, they have also pushed forward the frontier of our understanding of economic growth and development.
        
        important findings
        
        Among the notable ―revisions‖ made by cliometicians were the findings of Conrad and Meyer (1958), Yasuba (1961) and Sutch (1965) that slavery was indeed a profitable investment. Easterlin (1961) used revised GNP figures to show that income in the antebellum South grew at a faster rate than previously believed, and Fogel(1964) showed that the railroad was not the determinant of American economic development that it was believed to have been.
      - New approaches
        
        for the study of economics
        
        Finally, cliometrics has spawned entire new approaches to the study of economics. At the forefront are the new institutional economics, pioneered by Douglass North, and anthropometrics, which counts among its initial practitioners Robert Fogel. It is no coincidence that these two were recognized with the Nobel Prize in Economic Science in 1993.
        
        North
        
        North was an early practitioner of the ―new‖ economic history. In one of his most famous papers he measured the impact of decreased transoceanic shipping costs. His surprising finding was not that shipping costs decreased, which was widely recognized at the time, but that it was not technology, so much as institutional changes that were the source of the decreased costs - less piracy and quicker turnaround times. This emphasis on institutions ultimately defined North‘s career.
        
        Perhaps the most influential book to come from the new economic history is North‘s Economic Growth of the United States, 1790-1860 (1961). Though seriously lacking in through empirical research, its contribution ―lay in the vivid and concrete way it showed American historians how a large economic model, nonmathematical but theoretically sophisticated, could be used to organize the economic materials for the several American regions over half a century of economic history
        
        Norths early work
        
        In North‘s early work (1961 and 1966) he focused on the standard neoclassical explanations for economic growth (technology, human capital, technological change). But when he began to study European economic history he concluded that the neoclassical model was not able to explain the kind of fundamental societal change that had characterized European economies for the past 500 years. This led him down the path of what would become the newinstitutional economics, making him an early proponent of two different revolutionary schools of economic practice: cliometrics and new institutional economics.
        
        importance of institutions
        
        In a number of books, beginning with Institutional Change and American Economic Growth (1971, with Lance Davis), North demonstrated the importance of the role played by institutions (including property rights) on economic development. In Institutions, Institutional Change and Economic Performance (1990), he posed the fundamental question of why some
        
        countries are rich and others poor. His conclusion was that institutions are a major determinant in the profitability and feasibility of economic activity. The greater the institutional uncertainty, the greater the transaction costs and the greater the drag on economic growth and development. These views were a novel approach in both the history and development fields. Typical economic growth models focused on technological change and capital accumulation, assuming zero transactions costs and ignoring institutions altogether. He maintained that new institutions arise when groups in society see a possibility of profiting that is impossible under prevailing institutional conditions. If external factors make an increase in income possible, but institutional factors prevent it, then new institutional arrangements are likely to develop.
  - - - Clio has not had an uncheckered history. Its growing popularity has led to a rift between economists who practice cliometrics and historians who practice economic history without the use of the formal models, which they argue miss the context of the problem and have become too enamored of statistical significance at the cost of contextual relevance.
    - - Boldizzoni attacked cliometrics, focusing his sharpest criticism on the quantification of history at the perceived expense of its humanity.
    - - On the other side, cliometrics has lost some of its significance with economists, who see it as another application of economic theory, albeit using historical data. While applied economics is not seen as a bad thing, cliometrics is not seen as anything special. Rather, it is often perceived as the application of theory and the latest quantitative techniques to old data instead of contemporary data. In that world view, a cliometrician is just a theorist with a more limited repertoire – and hence a luxury in an environment of shrinking resources. As a result, cliometrics has been blamed to a degree for the demise of economic history positions in many economics departments.
      - William Parker
        
        As early as 1986 William Parker foreshadowed this problem when he observed that what was lost in the move to theory and econometric emphasis was the humane interest of the old British political economy and social welfare and the idealistic German historical economist‘s concern for the whole society.18 At the same time,
      - Alex Field
        
        Alex Field (1987) cited problems from another flank. Whereas the ―new‖ economic historians had to fight to prove their technical skills belonged in the study of history, by the late 1980s there were no more ―old‖ economic historians left to challenge. Instead, the challenge came from the other side, where economic theorists questioned what value cliometricians added to departments strapped for resources. Most economists possess the same or even more sophisticated technical skills, which can be applied to any data set, contemporary or historical.
  - - - So in the end, what is cliometrics and what is its place in the history and lexicon of economic history? Is it history with an economic (technical) approach? Or is it economics with a long run view of the world? Or has cliometrics become a subset of economic theory? The answer, not surprisingly, is all of the above
    - - For many authors and many of its protagonistscliometrics appears to be first of all a branch of history. Using economic tools, techniques and theories, it provides answers to historical, rather than contemporary economic debates.
      - inductive
        
        This inductive view is intimately linked with the German historical school, despite the use of more sophisticated techniques. It could be said that the two disciplines became closer, but probably within the frame of ‗inductive‘ economics. On top of that, despite those early interests in building a kind of historically (i.e. inductively) grounded development economics, c
        
        tried to provide answers to historiographical questions
        
        cliometrics mainly tried to provide answers to historiographical questions  and therefore spoke more to the historian than to the standard economist. Econometric techniques may be used, with the reconstitution of time series and identification of missing figures by interpolation or extrapolation a practice that professional historians seldom condone. But these cliometric procedures do have a historical vocation - that of shedding light on historical questions considering economic theory or econometrics as auxiliary disciplines of history. And when the cliometric approach was mobilised to build a development theory based upon clearly measured facts, it developed an economics more akin to the objectives of the German Historical School than one participating to the movement towards highly abstract and deductive theory that characterised the development of the neo-classical school of the time.
      - in short, two options
        
        option 1
        
        use of statistical techniques such as discriminant analysis) does not turn the discipline into a branch of natural science
        
        option 2
        
        or the cliometric approach is mobilised to obtain theoretical results grounded more on induction from collected time series than from a deductive explicit modelling exercise, i.e. economic theory that must be primarily founded on facts and a generalisation of empirical evidence. In this way it contributes to an economic science that is more related to the German Historical School than to the neoclassical perspective.
    - - But this is not the end of the story. Some recent work in cliometrics performed by economists reveals the possibility of a cliometrics that could also be an auxiliary discipline of economics per se. As such, it should be part of the toolkit and competencies of all economists. However, as the term auxiliary discipline indicates, it could only fulfil its proper role for economics if it remains slightly outside the realm of standard neoclassical economics. It must be a compound of the application of the newest econometric techniques and economic theory with the old institutional and factual culture characterising the old economic history.
      - the role of history
        
        History is indeed always a discipline of synthesis. It should also be the case for cliometrics. If not, if cliometrics were to be deprived of all its ―historical dimensions,‖ it would simply be economics applied to the past, and would thus cease to exist as a separate branch of study. To be helpful for the economics profession at large, its main contribution should be to mobilise all the relevant information that can be gathered from history to enrich or even challenge economic theories. And this relevant information should also include cultural and institutional development.
      - quantitative is poor qualitative?
        
        A conventional belief among economists is that ―qualitative is poor quantitative.‖ But could ―quantitative is poor qualitative‖ also sometimes be true? A big difference between economists and historians is the sense of so-called historical criticism and the desire to avoid any anachronism. In addition to close examination of the historical sources, this involves the close examination of the institutional, social and cultural context that forms the framework constraining the players‘ behavior. Cliometrics will not build a general theory because it shares too strongly in the belief that it is necessary to examine economic phenomena in their context. But it could suggest a few useful ideas and insights, based upon solid investigations and correctly estimated stylised facts, to economists who are attempting to develop laws of economic behaviour.
      - Summary
        
        In summary, it could be contended that a good cliometric practise is not an easy exercise. Becoming too narrowly ―economic,‖ it would not be possible for cliometrics to answer certain
        12questions that would require, for example, more information about the microstructure of financial markets or the actual functioning of stock exchanges during the period under scrutiny  it would only measure phenomenon that it cannot explain. It would require the specific approach (and extraneous information) of the historian to describe the reasons for the lack of relevance (or understand the shortcoming) of such an economic theory in a given context (precise place and period). It is perhaps only in this regard that cliometrics can provide something for economists by suggesting lines of research. However, if it became too ―historical,‖ cliometrics would cease to appeal to the economics profession. Economists need cliometricians aware of their debates and their interests.
    - - Last, but not least, cliometrics could one day be more than just an ancillary discipline of economics and instead become a full-fledged field of economic theory. There is indeed another possibility: viewing cliometrics as the science of the emergence of institutional and organisational structures, and that of path dependence. Economic history would use the old techniques of the discipline coupled with the state of the art arsenal of econometrics in order to reveal stylised facts about the efficiency of various institutional arrangements as well as on the causes and consequences of institutional change. It would help the theorist in developing a true theory of institutional change, i.e. one that at the same time would be general (serving the needs of policy makers today, for example) and theoretically solid (grounded on economic principles), while solidly grounded on empirical regularities as put forward by a joint economic and historical analysis. This analysis of institutional morphogenesis would be the true theoretical part of a cliometric science that would emancipate itself from its apparently purely empirical fate  being the playing ground of long run econometricians. It is clear that economists‘ desire for generality and their fascination for the mathematical science does not encourage them to pay too much attention to contextualisation. However, neo-institutionalist economists like North warn us to seriously consider institutional and cultural contexts.
      - contributions of economic historians to the field of economics
        
        Economic historians have contributed to the development of economics in many ways, combining theory with quantitative methods, constructing and revising databases, and discovering and creating entirely new ones. This has made it possible to question and reassess earlier findings, thus increasing our knowledge, refining earlier conclusions, and correcting mistakes. In addition, this field has added greatly to our understanding of economic growth and development, affording the economic historian the valuable element of time as a variable, which the traditional theorist does not enjoy. The use of history to examine economic theory has deepened our knowledge and understanding within fundamental areas of research as to how, why, and when economic change occurs. It is perhaps in this area where the greatest contributions of economic historians have appeared.
        
        ultimate conclusion
        
        By merging economic history with modern techniques, cliometricians have not ended economic history, but elevated it. The continuing evolution of technology has made a tremendous impact on the ability of cliometricians to handle ever larger data sets, share them with a wider audience, and access new data sets that previously took a lifetime to collate. In conjunction with the greater facility current economic historians have with econometrics, the future seems limitless. But as any good historian knows, predicting it is fraught with perils