Business Magazine

The Disputability of Macroeconomic Knowledge

Posted on the 11 November 2011 by Realizingresonance @RealizResonance


Photo courtesy of iStockphoto

I believe that issues in macroeconomics are inherently disputable. That is to say, measurement of economic variables and subsequent analysis of cause and effect at the level of an entire economic system in its totality is not physically possible, unless we have Laplace’s Demon for an economist, an entity with an omniscient view of the past, present, and future. This results in the problem of persistent ambiguity in our full understanding of an entire economic system. The ambiguity always leaves room for interpretation and room for debate. Reasonable, rational, well -intentioned people are going to disagree about macroeconomic theory, and there is no objective way to really settle the issue to everyone’s satisfaction. This dispute has root in the schools of economic thought, but it takes on a dramatic form at the level of public debate over macroeconomic policy. I think that the natural ambiguity which comes from attempting to explain the workings of an entire economic system, including its potential future course based on contingent policy choices, builds disputability into macroeconomic discussions.

This claim might lead one to believe that I am an economics cynic, nihilist, or novice, all of which I am not and don’t want to be, so let me try to clarify a few things about this first. I want to point out that I am employing my personal experience studying economics and doing economic analysis, which is certainly not at the PhD level, does not include any advanced econometrics or macroeconomic modeling, but does include a Bachelor of Science plus professional experience with data mining, statistical analysis, and predictive modeling. I feel confident in my grasp of economic principles, arguments, and evidence, though I have plenty more to learn, and I reserve the right to modify my future economic thinking. In addition, I still have strong views about how things work, and I often feel confident in both my theoretical understanding and causal assumptions regarding macroeconomic policy. I am not claiming that we can’t have good and expanding knowledge of economics leading to more informed and justified beliefs, just that reality is inherently disputable at the level of an entire economy in its totality. While my limited experience and confidence with economics might appear at odds with an argument for natural ambiguity, I contend that it really just helps me illustrate my thesis better. Regardless of how elite and experienced the economist, he always has plenty more to learn. Persistent macroeconomic disputes are obvious in politics, but they also involve the high lords of the economics profession and the business world.

For those who don’t know, the economics curriculum is generally separated into microeconomics and macroeconomics. The former studies economic principles and issues from the perspective of individuals, firms, industries, and particular markets, and the latter studies the economy as a whole and focuses on familiar metrics like GDP, employment, and inflation. Most undergraduate economics degrees provide a combination of micro and macroeconomics classes, including my degree which was also integrated with political science and philosophy. However, in many ways microeconomics and macroeconomics are themselves not well integrated. Most of the theory and evidence that I studied in microeconomics classes dealt with notions like scarcity, rational choice, competition, market principles, tradeoffs, cost-benefit calculation, marginal analysis, net present value calculation, investments, incentives, and most things that affect profits and losses, from the individual’s or the firm’s perspective. In macroeconomics classes I spent more time learning about systemic factors such as aggregate supply and demand, business cycles, money and banking, international trade, monetary policy, fiscal policy, and government regulation, from the perspective of omniscience. Much of the debate about economics in the public sphere sounds like arguments between microeconomics and macroeconomics to me, debates which loosely correspond to economic arguments from the right and the left of the political spectrum. I think there is an integration problem. The uncomfortable integration with microeconomics is a strong factor in the creation of ambiguity and the inherent license for disputability over macroeconomic policy.

Microeconomics and macroeconomics are clearly dealing with the exact same phenomenon of economics, but there are some significant differences in the perspectives and the conceptions. For example, the more micro one gets with economic principles the more solid they are, but the more macro one gets the more abstract economic principles necessarily get. Evidence is more abundant at the micro level, but scarce at the macro level, which is not always obvious since the macro evidence is an accumulation of all of the micro evidence plus any evidence based in aggregate statistics. But consider that with microeconomic theories there is opportunity for experiment, testability, falsifiability, and confirmability. The method of discovery for principles of microeconomics is not only closer to science it leverages intuition and common sense much better since the individual human mind participates directly at the micro level as consumer, worker, or entrepreneur. Micro analysis deals with smaller scale problems over smaller time frames. To discover macroeconomic principles one must attempt to synthesize all of the data everywhere into one coherent picture of the entire economy as it unfolds over time, and since this is virtually impossible to do completely, the shortcuts of aggregation and/or intuitive logic are employed for the task. But macroeconomic intuition is much more difficult than microeconomic intuition, being far from, and at odds with, the natural intuitions of any lone individual mind. I liken complete macroeconomic clarity to something like achieving Buddhist enlightenment, because you must transcend the individual ego and become one with the economy.

Two economists who agree on everything about microeconomics can easily find disagreement about the implications for macroeconomics. This is where economists can breakout into camps or schools, with New Keynesians, Monetarists, Neo Classicals, Supply-Siders, and the Austrian School being the most prominent segregations in the last few decades. During the Great Depression economists competed to explain why the downturn had become so persistent, a contest that saw the rise of John Maynard Keynes, who promoted new notions that dealt directly with aggregate phenomena and essentially began the separate discipline of macroeconomics. Keynesianism contributed concepts like animal spirits and the multiplier effect, plus the proposition of benefits from budget deficits, fiscal stimulus, and other tools for dealing with drops in aggregate demand and confidence. While these ideas were mainstream for some time, strong oppositions developed that sought to place some focus back onto microeconomic principles as the main drivers of macroeconomic patterns. Milton Friedman created the Monetarist School with the theory that changes in the supply of money were the ultimate determinant of business cycles and the facts about the Great Depression, and if money growth were steady then all that was left to understand was microeconomics. Neo Classicals have developed macroeconomic theories and models rooted in microeconomics principles, with Robert Lucas developing rational expectations theory, shining light on how individuals respond to macroeconomic policy. Supply-Siders, like Arthur Laffer, argued that macroeconomic goals were best served by government policies that encouraged production and supply, disputing the Keynesian focus on demand. I have not exhausted the contributions and disputes which originate from all of the different schools, but for my thesis, it is sufficient to know that there are various competing schools of macroeconomic thought and that their disputes don’t appear to have any imaginable resolutions that will lead them all to converge in the foreseeable future.

Austrian School ideas have been gaining popularity over the last few years, a phenomenon that I think can be attributed to GOP Presidential Candidate Ron Paul. Austrians focus on liberating the free market so that it can work its proper magic of coordination through the price system and incentives. In this view the business cycle develops busts and booms because of government intervention, which always results in distortions of the market mechanisms and subsequent malinvestment. Besides being an additional example of divergent views of macroeconomics, a central argument of the Austrian School contributes directly to my thesis, although I do not necessarily agree with their conclusions. The Austrians do a good job of demonstrating the intractable problem of acquiring complete empirical knowledge of an entire economy, but I don’t find their rejection of empirical induction in favor of logical deduction to be any more helpful in illuminating macroeconomic scenarios or proper policy choices.

Ludwig von Mises criticized the ability of central economic planning to work efficiently due to the fact that change was unavoidable in economic life, and abstract models for planning could never achieve collectively all of the information available to all of the economic players acting independently. Market information about scarcity and availability, and relative values established by competition are established by the price system. Without this information there is no way for governments and people to know what to produce and how to produce it most efficiently. Friedrich von Hayek argued that the profit motive and market mechanism allows people to use their disperse knowledge to create new knowledge, and that efficiency depends on utilizing this knowledge being used at particular times and places, something which could not be discovered through aggregate analysis of historical data from a central location. The Austrian School has rejected the use of aggregate statistics and econometric modeling as useful, and this bone of contention goes all the way back to the 19th century and a dispute between the Austrian Carl Menger and the now extinct German Historical School over epistemology, called the Methodenstreit. The Germans favored an historical and empirical method, while the Austrians favored the employment of rational extrapolations from pure principles. The inherent disputability in macroeconomics arises out of this ambiguity in epistemology, since it will always be a fact that historical aggregated statistics are a necessarily incomplete and flawed picture of the entire economy.

The Austrian School solution is to assume that since aggregated knowledge is imperfect and misleading, and since any act by government to intervene in the economy distorts the usefulness of employing dispersed knowledge, the best thing to be done is let the economy develop organically by free markets with minimal intervention through the enforcement of property rights. However, it is a significant leap of faith to assume that this will necessarily work out, that an entire economy is necessarily self correcting, that economies never evolve in ways that require novel governance and regulation, and that the most optimum outcomes will always be driven by free markets and libertarian governance. Nor is it obvious to me the use of statistical analysis in order to discover knowledge about macroeconomic scenarios is a worse method than rational extrapolation from principles. I have very direct experience with the usefulness of statistics, which I derive great utility from and develop new knowledge because of, and it really is the only game in town if you want to investigate the complex macroeconomic reality as it actually happened. Dispersed knowledge is useful for obtaining facts on the ground, but knowledge derived from historical analysis is useful too, and without it we would have even less total knowledge about the entire economy. Disputability and ambiguity about policy decisions is not solved by the logic of the Austrian School.

There are other problems with extrapolating from microeconomic principles. The fallacy of composition illustrates why we should not make a logical leap from the facts about microeconomics to macroeconomic conclusions. A group of individuals all acting rationally can cause irrational and suboptimum outcomes in the aggregate. If I stand up at a concert to get a better view, this may cause others to stand, and once everyone is standing the view has not changed for anyone despite the efforts of everyone in the audience. In game theory there is a problem called the prisoner’s dilemma which shows that acting in rational self interest results in a less the optimum outcome for the players. There is the free rider problem which illustrates the way that citizens can benefit from public goods while avoiding the personal costs. Then there are the problems of negative externalities like pollution, asymmetric information in the sale of used cars and insurance leading to adverse selection, the mania of bubble psychology, the contagion of financial panics, and the slow and uneven adjustment of prices. Not to mention the new field of behavioral economics that has shown economic decisions may often be totally irrational and driven by cognitive biases. Macroeconomics emerges from the totality of factors that affect microeconomic behavior, but the patterns that spring forth are not obvious and intuitive from the micro perspective.

Ceteris paribus is a Latin phrase which means “all other things being equal”, and economic laws and precise factual statements about generalized economic causes and effects must usually be qualified with this phrase. For example, the law of demand indicates that if the price of a good or service falls then the quantity demanded for that good or service will rise, ceteris paribus. The law of demand is behind the popularity of sales on retail goods, a common sense notion, and as far as I know there is not a serious dispute about the existence of this law in the economics profession. However, if you leave out the ceteris paribus and attempt to eternalize this law everywhere and always, it is easy to find counterexamples. Consider a beleaguered tech company that lowers the price of a new tablet PC, but because sitting on the shelf next to it is an iPad, the company’s sales continue to lag. Or that no matter how much a manager lowers the price on some previously contaminated food item the consumer knowledge about the contamination will prevent sales at any price. If you have been following the housing market over the last few years you will know that housing prices continue to fall at the same time that home sales also continue to fall, with an obvious connection between the two, and there are still serious problems getting new people into the market even with falling prices. Without ceteris paribus the law of demand is just a tendency of demand.

I am not criticizing the use of ceteris paribus by any means. It is a method of reduction, to explain cause and effect at the ground level, a tried and true scientific practice. If we break down an object or phenomenon into its constituent parts for careful analysis we can learn significant and useful things about how things work at the micro level, and then we can attempt to put the knowledge of the parts together in order to understand the whole object or phenomenon. Reduction in economics works well for understanding the laws and facts that elucidate simple abstract concepts which can explain isolated situations, resulting in useful information for many economic actors, whether in business, government, or as consumers. However, using the ceteris paribus facts about microeconomics to reconstruct the facts about an entire economy has consistently confounded the economics profession, and this contributes to the divergence of views over macroeconomics. I will suggest that where ceteris paribus ends, divergent schools of economic thought begin.

A common way to study macroeconomics these days is to use statistical and mathematical approaches to search for evidence in the historical record. Many Nobel Prizes have been awarded on the basis of innovation in methods of analysis, and the discoveries that are made with these methods. New Keynesians, Monetarists, Neo Classicals, and Supply-Siders have all discovered support for their theories in the analysis of historical data. There is a bit of dialectic in the mainstream of economics, whereby Keynesians have modified their views based on good evidence from the New Classicals, and visa versa. Yet these schools argue over abstract fundamentals such as general equilibrium theory and whether people have money illusion. Even more relevant to actual economies are the disputes over what to do about fiscal and monetary policy, decisions which are not settled by the various economic schools of thought or the evidential analysis that support their theories. Even if all economists agreed on macroeconomic theory and there were no separate schools, it is not likely that the collective knowledge of the historical record would be enough to stop politicians, business, and the public from contesting policy with reasonable arguments, especially since the total history of agreed upon economic knowledge will still necessarily have gaps, and the reality of economic conditions will always be outpacing the aggregate knowledge of it.

Economists attempt to make a distinction between factual statements, called positive economics, and value based statements, called normative economics. Examples of positive economic statements are things like the law of demand, which I mentioned earlier, or historical facts like the Great Depression began with the Great Crash of 1929. These are neutral in value. Examples of normative economic statements are often expressed as shoulds or oughts, such as the government should increase aggregate demand during a recession, or we ought to address the issue of income inequality. These are not neutral in value. Many economists see their profession as the most scientific of the social sciences and may try to avoid the normative statements, and this is easy to understand since numerical economic data is abundant due to the widespread public and private accounting of money, a dimension which allows for the application of advanced mathematics and statistical resources. The tools of professional economists can then be useful for various divergent purposes; the positive elements can be used for different normative outcomes. Economics is a social and political issue, so the abstractions of mathematic models will always need to be converted back into the social and political reality. For this reason, positive and normative economics are not really separable in most people’s minds, including economists. Facts may not be disputable, but values and opinions are.

Macroeconomic policy is not simply the concern of economists in an academic sense, but has massive implications for the real world, the public, and by extension politics. Positive economic statements are not sufficient for this wider economic discourse, since the best that these can typically do is shed light on the various tradeoffs in economics. The public constituency is not neutral to economic life, and they will expect that economic policy will be beneficial to them in some way, as it should result in employment opportunities, equitable incomes, stable prices, and the availability of necessary and desired products and services. Achieving normative macroeconomic goals is difficult and contentious, not only because there are multiple theoretical views to decide between, but even more so because, even with agreement on positive scientific statements about economics, there will still be disagreement on the normative goals of economic policy. Different groups are going to benefit more, or less, from different policies, and these disputes are hopelessly subjective for the participants. For example, if there is a tradeoff between inflation and unemployment, such that when unemployment declines inflation increases, then those who are already employed might justifiably oppose government programs designed to increase employment, if this would also mean rising household costs for them. Of course if you are unemployed then inflation is not your primary concern. This debate can be seen in the arguments over Federal Reserve policies like Quantitative Easing. Academic economists may hope for their discipline to remain positive and scientific, but the reality of economic life for regular people means that there must be tradeoffs made between economic goals, and these are always up for dispute.

Even when the public appears to agree on long term macroeconomic goals, there are still disputes over the best policies to achieve these purposes, including how to judge the effectiveness of past policy. It would be difficult to find people who are against reducing unemployment, or people who are for slowing down economic growth, but ask people about the effectiveness of President Obama’s stimulus plan and there will be vehement disagreement. The stimulus package in 2009 is often perceived to have failed, because it did not fulfill its publically stated purpose of holding the national unemployment rate below 8%, which was the goal Obama had set for the package. Many argue that the stimulus did nothing for jobs, and only increased the budget deficit. For others, including me, it is easy to explain why Obama’s forecast of 8% unemployment was incorrect, while also maintaining that the stimulus worked in the way it was intended. Obama and his economics team failed to forecast how fast and deep the economic numbers were going to fall, not how beneficial the stimulus package would be. The nonpartisan Congressional Budget Office (CBO) estimates that the impact in the stimulus on second quarter of 2011 resulted in GDP growth between 0.8 and 2.5 percent, lowered unemployment by 0.5 and 1.6 percent, and increased the number of full time jobs by 1.4 to 4 million. However, anyone, including economists, can disagree with the CBO, because in order to estimate the effects of macroeconomic policy, abstract models of what would have happened otherwise need to be built for comparison. It is not objectively possible to build an abstract model of something that did not happen, not everyone will agree. There really can be no scientific agreement on which counterfactuals to use, without appealing back to theory and this is unfortunately circular. I think the stimulus worked, but I don’t have a successful argument for anyone who wants to disagree with the counterfactuals that support my conclusion. Even if there were agreement on the counterfactuals for comparison of the short run effects of the stimulus, there can still be persistent disagreement about the long run implications.


Chart courtesy of The New York Times and Econbrowser Times

I have come across many people who unwittingly us the phrase, “it’s not rocket science”, to describe economic decisions when referring to matters of public policy. For example, someone might say something like, “if we raise taxes on businesses, then it will reduce the funds available for employment, and therefore cause higher unemployment; it’s not rocket science.” This is the simple solution fallacy. While the previous economic statement is generally not controversial in regards to the general economic theory that is being employed, namely a theory of labor demand, it leaves out the necessary ceteris paribus qualifications, which if included would limit the statements implications for the entire economy. A really large scale policy change, like a business tax increase, is not done in isolation, but within an open, dynamic, and adaptive system, so the implications for the entire economy are not obvious or even verifiable. Macroeconomics does not just involve the numbers that we can measure, but essentially everything that affects the behavior of people, and which the behavior of people effect, altogether, everywhere and always. So it is not rocket science, it is actually much more involved than that.

Rocket science is complicated, but macroeconomics is complex. While complicated things can be difficult to grasp, they deal with finite, closed systems, with variables and dynamics that are constrained and predictable in principle. Complicated things might be difficult to comprehend, but they are comprehensible, and once understood there is little ambiguity. Complex things are infinitely more difficult, because they deal with open non-linear systems within systems, with rules that can transform without warning. We can never have complete information on the factors that affect open complex systems, which creates an inherent uncertainty, incalculable risk, persistent ambiguity, and generally confounds any long range prediction of the future changes within the system. An entire economy is hopelessly complex and is always undergoing changes, both mundane and novel. So macroeconomics is incalculably more difficult than rocket science. Isaac Newton, the natural philosopher who invented calculus, the mathematics that rocket scientists still use today, is reported to have said that he could, “calculate the motions of erratic bodies, but not the madness of a multitude”, after losing a fortune in the 1720 South Sea Bubble.

My skepticism about the resolvability of macroeconomic disputes does not mean that I think that there is no hope for the objective economic knowledge. Nor does it mean that I do not have strong beliefs about what we should do. It just means that I recognize that no one is in possession of macroeconomic omniscience, and there are silver economic policy bullets. I often find myself disagreeing with the professional economists I see in the media whom I clearly have less knowledge and experience than, and there are concerned citizens on the internet, whom I clearly have more knowledge and experience than, but who find it easy to disagree with me about economic policy. I just cannot imagine the emergence of a new study or theory that ultimately settles these disputes. As Socrates indicated long ago, to be wise about something is to recognize that you are not in possession of unequivocally perfect knowledge about the things that you think you know, no matter how expert. To be wise about macroeconomics means being humble in regards to economic knowledge and how it is woven together. The more I learn the more humble I get. This makes it all the more frustrating to hear Presidential candidates, media personalities, and various citizen movements, all proclaiming with certitude their knowledge of the best macroeconomic policies. More than ever we need to recognize the disputability of all our macroeconomic claims, and embrace the humbleness of uncertainty. More than ever we need moderation in our economic views.

Jared Roy Endicott

The Disputability of Macroeconomic Knowledge
 Subscribe in a reader

Back to Featured Articles on Logo Paperblog