Or, How to adapt the intermediate macro syllabus to an altered world
This semester is the first time I’ve taught intermediate macroeconomics link in over two years. The last time I taught this course in the Spring of 2007, the key topics were inflation, the possibility of stagflation, and the possibility of containing the ongoing housing slowdown.
As I started the semester, I knew that I’d have to alter the emphases, and be ready to account for developments. The two big concerns were (1) dealing with the Taylor rule, and (2) dealing with the banking sector. A less difficult-to-deal issue is the consumption function.
Figure 1: Source: Deutsche Bank, Global Economic Perspectives, December 9, 2009 [not online]
Figure 1 highlights the first challenge. Over the past ten years, the trend in macro textbooks has been to dispense either partly or fully with the IS-LM construct, where the quantity of money enters into the determination of GDP, and substitute in a monetary reaction function, where the arguments are the output and inflation gaps, i.e., the Taylor rule. This was a useful innovation, but was difficult to apply to Japan (as I stressed in my lectures) and as of late 2008 as the zero interest bound became a reality for American policymakers.
The problem is even more obvious once one punches in the numbers and obtains the implied Fed Funds rate (below is the St. Louis Fed estimate, based on the CBO output gap, and differing inflation targets for PCE inflation).
Figure 2: Source: St. Louis Fed Monetary Trends.
As is well known, using plausible estimates of the output gap, the implied Fed Funds rate is negative. It’s at this point that one has to start waving one’s hands, and talking about quantitative easing…
So, IS-LM still seems to be a useful construct for analyzing fiscal and monetary policy efficacy. The IS-LM framework — and more broadly the Aggregate Demand-Aggregate Supply (aka “NeoClassical Synthesis”) — can accommodate the fiscal policy pessimists (as I discuss in my post here), and (of course) the liquidity trap critique of monetary policy (keeping in mind Joe Gagnon’s point). Of course, conveying the idea of credit easing is still hard to convey in this simple construct.
Which leads me to challenge number (2), namely how to incorporate different aspects of the financial sector. I chose to use a fairly simple framework, namely the Bernanke-Blinder CC-LM model (discussed in this post). The exposition is here; this is essentially a linearized version of the CC-LM model. The comparative statics are fairly straightforward.
While the CC-LM model heightens the amount of complexity, it does provide a rationale for Fed actions in terms of quantitative easing shifting out both the CC and LM curves. (On the other hand, it does highlight the fact that payment on reserves has limited the outward shift in the two curves). Of course, even incorporating the bank sector doesn’t mean that all the channels for monetary policy are accounted for (see this post for the others).
Finally, it is of interest to note that a discussion of the prospects for US economic growth, and consumption behavior, cannot be realistically appraised using the standard Keynesian consumption function. Rather, some discussion of how consumption depends on both current disposable income and net household wealth is essential.
Figure 3: Log real consumption, in billions Ch.2005$, SAAR (blue, left scale), and log real household net worth, in billions Ch.2005$, deflated by PCE deflator (red, right scale). Source: BEA, 2009Q3 2nd release, and Federal Reserve Board Flow of Funds, December 10 release.
Next semester, I’ll be considering how to incorporate the longer term implications of the Great Recession. One important macro factor involves the implications of the financial sector turmoil for the capital accumulation, and unemployment and the decline in asset values for labor force participation rates. In my seminar on the Great Recession, I discuss the recent OECD Economic Outlook Chapter 4 on this subject.
Of course, in adding new material to the undergrad course, some material has to be dropped. In my case, I reduced the time devoted to the New Classical models (the Lucas supply curve, and Real Business Cycle models).
Well, here’s how I’d start:
“Economics is about optimization, in particular the optimal use of scarce resources. But to what end? Since at least the French Revolution, three competing and durable value systems have characterized western culture, that is, liberte, egalite and fraternite. These are three objective functions, three different things we are trying to maximize, and historically they have conflicted. It’s for that reason that these values are typically embodied in political parties.
Egalitarianism is based upon the declining marginal utility of wealth and income, and it tends to lead to redistributive policies. The focus tends to be on the short term.
Liberalism–classical liberalism–is about the maximizing the property rights of the individual, and it’s about voluntary interaction. Liberals are the ‘principals’ in principal-agent theory.
Social conservatives are about stability and risk minimization. If egalitarians are about the present and liberals about the future, conservatives are about the past. They focus on increasing returns to scale of human interaction, that is, they are about the corporate man, about the family, God and country. In principal-agent theory, they are the agent.
So we have three, largely irreconcilable objective functions. We economists pride ourselves on our analytical skills, on our objectivity and our intelligence. But, by and large, our tools are limited to optimizing one objective function at a time.
And that means our advice tends to be highly politicized. That is an important reason why, after so many years, professional economists cannot agree on either the appropriate metrics or the suitable action with which to handle economic topics like recessions, deficits, free trade or health care. In many cases, we’re not optimizing the same objective functions.
So, with that, we can turn to the great analytical thinkers and key tools in our profession. But, in the back of your mind, remember that we talking about the optimization of given values, and that values are what matter most.”
Very nice, thank you Menzie, will help me rethink this section for my next course
“Next semester, I’ll be considering how to incorporate the longer term implications of the Great Recession.”
I have the luxury of not having to teach macro these days. If I had to, I’d probably start by making two points.
1) We have to judge our estimates of the longer term impact of recessions against our overall uncertainty about long-run estimates of potential output, its growth rate, etc. For example, a projected drop of 4% over the next few years sounds more serious than a forecast for a drop of 4% +/- 12%.
2) There’s a lot of evidence that business cycles are asymmetric; shocks in deep recessions seem to have very different long-run impacts than shocks at or near the top of business cycle. Quite different modeling approaches give the consistent view that shocks in recessions are much less persistent.
Point #1 is hard for me to cope with. I don’t know where I can find credible estimates of the degree of uncertainty of long-horizon estimates of potential output or its growth rate.
Point #2 makes me skeptical of the many studies that assume that shocks have the same impact in recessions and expansions. I’d start by dividing the empirical results into those that do and those that don’t.
Once I get past those two point, most of my recent thinking has been about the costs of financial bailouts on long-run growth. (I’m a finance professor….I can’t help it.)
If the government takes over troubled institutions, their losses get funded from future income taxes, which imposes a deadweight loss. The alternative is to let these institutions recapitalize themselves from future profits. My conceptual problem is that the latter seems to me like tax on capital. I would have thought that this should have higher deadweight loss than taxing income. Furthermore, economists I read who feel most strongly about the destructive effects of taxes don’t seem to mind higher banking sector profits as a means of rebuilding bank balance sheets. I don’t understand how they are thinking about these profits.
I see nothing about too much debt.
The economists who got it right focused on debt, the ability to make interest payments, and high asset prices.
How about this question? If there is already too much debt and there is an output gap, what should be done?
Menzie,
Before you decide what to put in or take out of a course curriculum, don’t you have to first decide whether you want to emphasize how an economy gets into a mess versus how an economy gets out of a mess like the one we’re in today? Focusing on the credit and banking part makes sense if you’re into autopsy mode, but focusing on the role of fiscal policy in a liquidity trap (along with the whole multiplier debate) probably makes more sense if you’re into prescription mode. So if you want your students to study the corpse, talk about banking and credit; but if you want your students to think about policy going forward, then I’d dust off that old IS-LM framework as a good way to convey the right intuitions.
Here’s the model I use. It’s simple, incorporates risk premia on financial assets, and gets right to the heart of the zero bound problem.
http://www.gettysburg.edu/dotAsset/2104335.pdf
Thanks to Menzie for this is very useful post. I haven’t taught intermediate macro for quite some time (the last time I did, I used, as Menzie does, Robert Hall’s text, but then it was co-authored with John Taylor), but if I were to do so soon, the discussion of how the Bernanke-Blinder CC-LM model was incorporated into the course would be very helpful. On Simon’s point #2: it certainly would be beneficial if Beaudry and Koop’s result reported in their 1993 JME article that “the effect of a recession on the forecast of output is found to be negligible after only eight to twelve quarters” proves to be robust in the upcoming portion of the out-of-sample period.
Menzie wrote:
“In my case, I reduced the time devoted to the New Classical models (the Lucas supply curve, and Real Business Cycle models).”
My department has never allowed me to teach macro despite the fact that it is my field. I’ve always suspected it is because of my open hostility towards New Classical models (especially RBC), which if I had my way would be covered by treating it as a historical curiousity, making few snide jokes (a la Lucas) and then moving on.
P.S. For example: “Have you heard the one about the unemployed person? She was simply displaying an increased preference for liesure.”
Thanks to Menzie for opening a keg of worms and for Anonymous for the reminder that only one objective function can be mazimized at a time.
The politician who yearned for a one-handed economic advisor does not know how to use economic advice. For a sample of the kind of comprehensive even handed advice politicians should receive, see the current David Brooks article (in today’s NYTimes)
Modesty and clarity about the value basis of any discussion would be helpful.
If you are going to concentrate on the IS-LM model, please include Sir John Hicks recanting, or at least massive caveat that IS-LM was appropriate only in very limited circumstances. Usually Keynes is congealed to this mathematically convenient form, losing the nut.
Menzies says he’s going to drop RBC and Mark says he’s not allowed to lecture on them because he thinks they are extremely silly.
But there’s real business cycles and there’s real business cycles. After all, Menzies distinguished co-blogger James Hamilton is a bit of an RBC guy isn’t he (witness his 1988 JPE article). And Beaudry and Portier have models of TFP expectations shocks that are non-monetary.
And turning now to the Bernanke-Blinder model…doesn’t that have a money multiplier in it? Seems a bit archaic in a world of endogenous money.
Menzie,
If you are really interested in teaching rather than indoctrination you might offer an alternative reading list to your students for information and education. I realize that you would not be able to teach these.
Here is a short list of suggestions. I am sure there are others here who could offer other suggestions.(Many of these are available on line)
Ludwig von Mises
Socialism
The Theory of Money and Credit
Human Action
Lionel Robbins
The Great Depression
Friedrich von Hayek
The Road to Serfdom
The Constitution of Liberty
The Fatal Conceit
Frederic Bastiat
That Which is Seen, and That Which is Unseen
He Who Has a Right to Work Has a Right to Profit
The Law
Henry Hazlitt
Economics in One Lesson
The Failure of the New Economics: An Analysis of the Keynesian Fallacies
Benjamin Anderson
Economics and the Public Welfare
Jean-Baptiste Say
A Treatise on Political Economy
John Stuart Mill
Essays on Some Unsettled Questions of Political Economy
Jude Wanniski
The Way the World Works
Brian Domitrovic
Econoclasts: The Rebels Who Sparked the Supply-Side Revolution and Restored American Prosperity
Thomas Sowell
Says Law: An Historical Analysis
Basic Economics
Applied Economics
Economic Facts and Fallacies
Milton Friedman
Free to Choose
Shouldn’t economists teach the 1920’s, or is that in the history department?
“If you are really interested in teaching rather than indoctrination you might offer an alternative reading list to your students for information and education”
RicardoZ either wins the 2009 award for stunning lack of self-awareness or gets a special exemption in the healthcare bill for extracting tongue from deep in cheek.
Either way, well-played my lord.
Cedric Regula,
Economic History is no longer taught at the graduate level, in large part to Neo-Classicist/RCB types. I personally think it is a huge mistake.
The first thing I would do, Professor, is go read what Alan Kirman
has to say about the implications of Sonnenschein Mantel Debreu.
It demolishes the link between micro and Macro. And the Kirman
piece is extraordinarily well written, and of modest length. Enjoy
Menzie,
I forgot this one. It is a good one to stir thinking from your students.
Robert Bartley
The Seven Fat Years: And How to Do It Again
Menzie,
A cruious question came to me and it is one you might ask your students and answer yourself.
We know that supply and demand exist in any economy, but would demand side economic theory (Keynesianism and monetarism) exist without money and government?
Mark A. Sadowski, I would love for you to take a shot at answering this.
Cedric, I would also be interested in what you would say about this. You seem to be more thoughful.
Ricardo,
I can’t really take credit for being more thoughtful on the subject. My real secret is I’m an avid fan of sci-fi and therefore have experience with far more socio-economic systems, spanning many more planets, civilizations, and alternate universes, than the average economist with merely a PhD in Earth Economics.
But to answer your question, of course not. Demand is intrinsically what demand is, and demand side economics was created by the advent of the ability to create money and more importantly credit at the direction of what government thinks is proper.
Unfortunately, there are very few instances of civilizations that exist without money and credit, and none that exist without government of some sort.
So the devil is in the details.
I’ve found it useful to include Ch. 5 “The Reality of Macroeconomic Structures” from Kevin Hoover’s Causality in Macroeconomics (2001) in my reading package for Intermediate Macro (and Applied Econometrics). It helps motivate why a macroeconomics course should be partly a course in the history of economic thought and not just a summary of the latest “consensus” model. Reality determines the model, not vice versa.
RicardoZ,
That’s a very thought provoking question, to which I’m not sure I have a satisfactory answer.
The problem is it’s very hypothetical because, to my knowledge, no large scale barter economy has ever existed. And partly due to acculturation, when people are thrust into a situation where they are without a unit of account and a store of wealth they quickly improvise one. For example “The Economic Organization of a P.O.W. Camp” by R.A. Radford (Economica, November, 1945) describes how WW II POWs used cigarettes as a currency. They were homogeneous, reasonably durable, and of convenient size for the smallest or, in packets, for the largest transactions. And the economy was of course subject to inflation, deflation and “recessions” as the supply of cigarettes varied.
One way of thinking about your question is by considering the differences between the Neo-Classical and the Neo-Keynesian macroeconomic schools of thought. A fundamental difference concerns the ability of prices to adjust. If prices are highly flexible then business cycles should be primarily “real” or supply side in origin. If prices are “sticky” business cycles that are primarily “nominal” or demand side in origin can occur. An economy without a currency you would by definition have no prices so it’s hard to imagine demand side shocks to that economy. So I would say no.
RicardoZ: A standard way Keynesian models can be taught is to first describe the workings of the model with no government sector, in which case consumption and investment demand determine GDP.
I’m with SvN: the deadweight loss of recapitalization (especially now that the government’s equity participation in such recovery is being treated as something we should renege on) is going to far exceed, in damage to the economy, than a straight takeover would have.
Part of this is because the costs will be paid as part of the recovery, having a damping effect on growth. The other part is that the benefits of the financial sector to economic growth are, as Paul Volcker noted, more limited than that sector’s current share of GDP.
Menzie, thanks for the commentary, I was a student of the intermediate micro ¯o courses in mid 80’s. Things have changed, but not the comment regarding the analysis of consumption on the Keynesian analysis. Business and Consumer behavior is dictated on the ability to evaluate outcomes. The branch of Government, Legislative, is not doing the Federal Reserve any favors by their poor fiscal record of responsible governence. Therefore, the outcome is for existing businesses to ratchet down and meet IBT through fiscal management rather than growth. Furthermore, the opportunities of new business growth is uncertain given the current climate in Washington that makes the ability for decision making uncertain at best.
Bring back Milton Freidman
We know that supply and demand exist in any economy, but would demand side economic theory (Keynesianism and monetarism) exist without money and government?
Cedric – But to answer your question, of course not. Demand is intrinsically what demand is, and demand side economics was created by the advent of the ability to create money and more importantly credit at the direction of what government thinks is proper.
“What government thinks is proper.” Great answer! You lived up to my expectations.
Mark – The problem is it’s very hypothetical because, to my knowledge, no large scale barter economy has ever existed.
Hypothetical or not if you do not consider a barter society you have avoided the question.
JDH – A standard way Keynesian models can be taught is to first describe the workings of the model with no government sector, in which case consumption and investment demand determine GDP.
While the circular flow theory can be proposed without government I still believe you beg the question without explaining how damand side theory can be implemented without some authoritarian force of implementation (government). And to the issue of money can the circular flow theory be explained in a barter society?
RicardoZ,
How about you create the first large scale barter economy. Then economists will have the opportunity to study it and then compare. Until then…
Mark,
Until you can conceptualize a barter economy you cannot understand economics. Until you can conceptualize a barter economy you cannot understand third party exchange and money as a medium of exchange.
How about teaching some of the practicalities of how monetary policy (implementation) works, Menzie? By that I mean issues such as open market operations, standing facilities, interest on reserves etc. Few people cared about this before the crisis, and my guess is that the textbooks have yet to be updated in the light of the importance these operations assumed from 2008 onwards.
How ’bout some discussion of the Ponzi Economy?
i.e. how politicians of both parties rely on ever-increasing private and public leverage (debt) to pull demand forward to try to keep this thing propped up for one more election?
I guess everyone in the country should be required to retake Macro every 4 years, when the “correct” theories change as new data come to light!
LOL
As a physical scientist, I would suggest more consideration of: (1) uncertainty ranges and systematic errors in macro measurement methods… (2) feedback and change processes in complex systems described by chaotic mathematics… leading to (3) carefully considering whether *any* given mathematical macro model has *ever* actually been verified with long-term real-world data… and thus (4) the utter failure of macroeconomics as a “predictive”, “scientific” discipline – as evidenced by its failure to produce and train economists who can actually predict anything useful before it happens.
What Macro *appears* to be good at, is creating generations of economic leaders who generally all think alike, thus making it easy for anyone who understands their biases to game the systems that they manage. Which is exactly what the financial sector wants the Federal Reserve to do — manage the financial system in a predictable, game-able fashion…
Speaking as a quasi-Georgist (converted rather late in life), if I were teaching I’d emphasize the core role Land plays in our economy, both as a source and sink of wealth.
From the Fed’s Flow of Funds report, Table L.10, Line 21, Mortgage Debt of the Private Sector, in billions:
2000 $5,093
2001 5,643
2002 6,397
2003 7,200
2004 8,224
2005 9,325
2006 10,377
2007 11,070
2008 10,972
3Q09 10,828
Over the four years from 2003-2006, the private sector borrowed an additional four trillion dollars against their properties. Very little of this was secured by new production. $1T/year @ $50K per job is 20 million jobs being directly or indirectly supported by this flow during this time, putatively speaking.
Going forward, I think housing rents and corporate quasi-rents are going to be increasingly squeezed as the consumer find his paycheck going less and less further every week, and pay rises if not employment itself in doubt.
We so easily forget that the dominant expense in nearly everyone’s budget is paying for land use titles. These cost $50 or so to create, yet we all pay hundreds of thousands for them now.