Monday, November 30, 2015

300 years of interest rates

In this post I just did a bit of educated guessing about where the monetary regime breaks should go over the past 300+ years in the UK (time series from here). Turns out all but one (in 1790) roughly correspond to points where interest rates changed from being/not being "pegged" (see here, here, and here):

I plan on doing a future post where I see if I can make this into a more scientific proposition rather than some vague empirical eyeballing ...

Principal component = information equilibrium model?

John Cochrane has an interesting paper/blog post about forecasting interest rates. I'm not sure I've absorbed it all quite yet, but I have a quick take.

The key point Cochrane is making is that the reason adding an inflation term to forecast models of interest rates improves them is really just because inflation has a trend -- a trend that roughly follows the first principal component term (PC1 at the link). Adding a trend (with principal components) allows you to get a really good fit -- and in fact it is this trend that captures most of the forecasting capability of the model. Cochrane says this means there a strong one-factor model of bond yields across all maturity. Basically, one interest rate describes them all pretty well.

This is just a cheesy overlay on the principal component graph, but that first principal component seems to be well described by the information equilibrium model:

Sunday, November 29, 2015

Maximum entropy better than game theory (again)

Today Nick Rowe mentions the dictator/ultimatum game (I choose to divide a pot and if you refuse the division, we both get nothing ... or the dictator version where you get no input). It's another case where the maximum entropy guess is better than game theory. Game theory says the solution is 99.9% (or more) for the dictator and x = 0.1% (or less) for the other person if people were truly rational. Maximum entropy guesses x ≈ 50%, but allows x ≤ 50% if information transfer is non-ideal. It also would guess x = 33% for three players, x = 25% for four, etc.

More on the rest of Nick's post later, but it brings up this again. And this.

Saturday, November 28, 2015

Department of Huh? Rent control edition

From Wikimedia Commons.

I got a couple paragraphs into this post at Project Syndicate by Brad DeLong and hit a stumbling block:
The reason that rent control is disliked is that it forbids transactions that would benefit both the renter and the landlord. When a government agency imposes a rent ceiling, it prohibits landlords from charging more than a set amount. This distorts the market, leaving empty apartments that landlords would be willing to rent at higher prices and preventing renters from offering what they are truly willing to pay.
Huh? DeLong is usually quite good at this sort of thing, so it's possible I'm missing something.

Why would a landlord choose zero rent and an empty apartment over renting at the rent ceiling? It's possible upkeep costs more than the rent ceiling with a renter rather than leaving the apartment empty, but I thought this was Econ 101 analysis.

And why would getting an apartment at the rent ceiling rate be worse for the renter than getting an apartment at the market rate? Sure Veblen goods come to mind (a thousand dollar a month NY loft apartment isn't as 'impressive' as a ten thousand dollar one ... to those who care about such things), as well as the ability to outbid someone else for the apartment.

I was under the impression that the reason price ceilings were bad in the Econ 101 sense is that they discouraged the building of new supply and acted as an implicit subsidy (basically producing shortages). If you can only get five hundred dollars a month for a one bedroom, it doesn't necessarily make as much sense to build a new building as it would if you could get a thousand. But additionally, more people are able to afford apartments, so fewer would be available. The result isn't empty apartments, but no apartments. And that's also what Wikipedia says.

So what does Info Econ 101 have to say about price ceilings?

In the case of a price minimum p0, there's a pretty simple solution. You basically restrict the price to p ∈ [p0, ∞) rather than p ∈ (0, ∞). However as the two spaces  (0, ∞)  and  [p0, ∞) are diffeomorphic to each other (except for the single point p0), you really end up with the same solution, just shifted. We can ignore the tiny detail of the single point at p0.

Taking the price to have a maximum value p0 is basically restricting the price to the domain (0, p0], which is more complicated. However there is a great function for mapping the space  (0, ∞) to (0, p0) where we'll again ignore the detail of including p0. We'll use an arctangent map so that the differential equation for the price with a price ceiling is the same as the differential equation without a price ceiling after the map arctan: (0, ∞) → (0, p0). This introduces an extra scale parameter (the "width" of the arctangent transition), but it can essentially be absorbed into the information transfer index. Now there are three regimes for the price ceiling. The first is where the ceiling is well above the equilibrium price:

The dashed line represents the demand curve with no price ceiling and the solid curves are the supply and demand curves with the price ceiling. For small shifts, this case is not very different than the original solution where the equilibrium price is marked with a black dot.

If the ceiling is comparable to (and below) the equilibrium price, you get a new equilibrium that is below -- but close to -- the ceiling:

And finally, if the price ceiling is well below the equilibrium, you get the standard Econ 101 result where the price is basically at the ceiling and the market is unresponsive to supply or demand shocks:

Update 11/29/2015

I just want to add that I am not advocating rent control, but rather the naive Info Econ 101 approach to understanding the effects that is more well-defined than the naive Econ 101 approach -- best would be a much more complex model.

Generally, my intuition is that a subsidy (or progressive taxation) would probably be the optimal approach. But it's also just a complex problem since space/land is genuinely limited.

Non-deflation non-surprise

Why didn’t the sustained high unemployment after 2008 push us into deflation? There are some popular stories — downward nominal wage rigidity that makes the long-run Phillips curve non-vertical at low inflation rates, “anchored” inflation expectations — and I cite those stories myself. But standard discourse on macroeconomics has not fully taken the non-deflation surprise into account.
That was Paul Krugman in his post from today.

Non-deflation isn't as much of a surprise in the information equilibrium model. Over time the response to inflation from output shocks has become approximately zero. And it's related to the liquidity trap -- the lack of a response to the price level from monetary expansion.

Thursday, November 26, 2015

The minimum wage in Info Econ 101

I added an update to this post about using supply and demand diagrams for the minimum wage, but thought it would be good to do a post on its own. The traditional Econ 101 use of supply and demand diagrams looks like this:

If you set a price above the equilibrium price P* then the new equilibrium market solution becomes the intersection of the demand curve and the minimum wage and the difference between the intersection of the demand curve and the supply curve with the minimum wage value is the amount of unemployment.

Some questions arise:

Why the demand curve intersection point? Why not the supply curve intersection point? This basically assumes that employers are at their limits, rather than say people are staying out of the labor force because a minimum wage job at a lower minimum wage isn't worth it.

What do the curves below the minimum mean? These represent the "illegal" supply and demand for labor -- the desire to employ people at less than the legal minimum wage. I personally would like to drive well over the speed limit, but that piece of the solution space doesn't necessarily enter into my speed decisions that are more based on traffic and road conditions. I'd really like to go a speed that is impossible for my car to attain ... but that should have even less impact.

Anyway, the Info Econ 101 (or Info Econ 1100101, ha! there's a nerd joke for you) picture is different:

The original solution to the information equilibrium condition is no longer a solution in the case of a minimum wage. The intersection point with the minimum wage and the original demand curve (or supply curve) is no longer a meaningful point. The general information equilibrium solution just says both supply and demand go up ... at least in the simplified presentation. There could be all kinds of more complex factors going on.

However that is the point. The Econ 101 analysis is not just oversimplified but wrong according to data. The Econ 101 view above is also wrong in the Info Econ 101 view which is itself simplified, but isn't inconsistent with data.
Econ 101: simple but inconsistent with data
Info Econ 101: simple and consistent with data

Wednesday, November 25, 2015

Speaking of math ...

... maybe you should prefer to use the IT model for quantitative predictions of things like inflation.

The new core PCE data for October says 0.1% inflation. The error is looking pretty good for the IT model as well (the IT model is doing as well as a model that consists of a smoothed version of the data -- gray dashed line):

David Beckworth's inflation corridor

I mentioned David Beckworth's corridor in comments on an earlier post and realized (again) I hadn't updated the graph from here. This one correctly plots quarterly data (instead of monthly) along with quarterly error (using the IT model prediction for core PCE inflation listed here; see links for model details).

Here you go:

Beckworth's model does show where QE1 and QE2 take off, but might be construed to have predicted QE4 at the end of 2014 -- which did not happen.

Looks like next year or two we'll start to get some real action (in the sense of which model is better) on this one. Basically, the IT model plus Beckworth's model (they're not inconsistent with each other if you don't take QE-n to cause anything) predicts a high likelihood of QE4. Considering how the Fed seems anxious to raise rates, this could be interesting.

Math up

My view of why you should include math is best captured by Paul Krugman:
My point is that there seems to be a lot of implicit theorizing going on here — and at least at first glance, the implicit theorizing doesn’t make a lot of sense. I could be wrong, but that’s the whole point of simple models: to lay bare what you’re assuming, and make it clear what, specifically, is driving your conclusions.
Krugman says "models", and that's what I mean by math -- some sort of formal construct that organizes the various factors you are trying to explain. It doesn't have to be a precise DSGE model. Generic supply and demand diagrams are math. Working with orders of magnitude and asymptotic behavior qualify as math. Basically, math lets other people use your model.

Math is the basic universal language for talking about how things relate to each other. If you are saying X is related to Y, you are implicitly using math. If you don't want to use math, you're not allowed to say X is related to Y.

Of course, some people think math doesn't help illuminate anything. In its most dogmatic form, we can see the opposition to mathematics from Ludwig von Mises:
All of mathematical economics, with its beautiful curves and equations, is idle flirtation. The setting up of equations and the drawing of curves must be preceded by nonmathematical considerations; the setting up of equations does not broaden our understanding. Mechanical equations can be used to solve practical problems through the introduction of empirically acquired constants and data; but equations of mathematical catallactics cannot in the same way be of service to practical problems in the area of human action where constant relations do not exist.
This is of course completely disproved by the success of the theories of how auctions work. As well as the information equilibrium model ...

But in general the reason for mathematics is not that humans are mathematical, but that economists' models of the macroeconomy are mathematical -- whether they want them to be or not. When you start saying effect E has cause C, then you are assuming a functional relationship between E and C: E = f(C). From that we can say that if C is small, we have E ~ f(0) + a C ... and we're off to the races. I've just turned your statement of C causes E into a linear model with an equilibrium value E0 = f(0) and an "elasticity" a. If you say "but a and/or f(0) changes" ... then is it really C that is causing E? Note that linear function E ~ f(0) + a C has exactly the same form if E = f(a) where C is the elasticity. And f(0) changing means that something else is causing E [1].

The reason people don't use math is either a) they can't think clearly enough to use math or b) they want to pull the wool over your eyes. The stated defense is that math doesn't clarify (that's false; it does) or that human actions don't have constant relations (my question to von Mises would be "then why even write anything ... once written down as math or words, human actions will change"). If human actions don't have constant relations, then what does it matter? Even history is studied because people think certain things can cause e.g. wars in a fairly consistent way like competition for natural resources. People study WWI and WWII because they think the things that happened then could have benefit in future conflicts (preventing wars or fighting them). That is the entire point behind "history repeating itself". If history never repeated itself, then there wouldn't really be any reason to study it.

Not using mathematics for your model:

  • Puts an upper bound on the complexity of your model (makes it easier to compete with smart people)
  • Allows you to say what you're saying is novel even if it is equivalent to existing models (allows you to fool people into thinking you have new insight)
  • Makes it impossible for anyone else to use your model (limits the use of your model to yourself)
  • Allows your model to change in respose to new information without saying you are changing it in response to new information (so you can fudge any data)
  • Allows you to get around stating clear predictions and precise conditions on those predictions (so you can fudge any prediction)
  • Allows you to stay away from quantitative statements about the data (so you can fudge any data)
  • Allows you to have effects of different magnitudes from causes of the same size (and vice versa) (so you can fudge any data)

A lack of math is not depth of argument or intuitive understanding, it is flim flam.


Update 7 September 2016

I should really have rephrased the part about "history repeating" in terms of "lessons of history" instead. People tend to take historical analogies too literally sometimes. An historical situation may enlighten you about the possibilities in a current situation (lessons), but there is almost never an exact analog of some historical event (repeating).



[1] This is my opinion of those expectation operators in models. An E[u(c)] term may look like your model depends on consumption or utility, but really it just depends on E.

DSGE form of the IT model (active but not interactive)

Well, the original idea of this post was to have an interactive version of the DSGE form of the IT model:

Oh, well. Here is the original source for the graph -- it's showing the effect of a monetary expansion (m, yellow) on output (n, blue), inflation (π, green) and interest rates (r, red) depending on the value of the information transfer (IT) index. This is short run (DSGE form is log-linearized, so is only true over a few time steps).

Tuesday, November 24, 2015

DSGE form of the IT model (interactive)

Testing some Wolfram cdf. This is an interactive implementation of this post that only works if you have wolfram CDF player installed and use Internet Explorer or Firefox.

Testing cloud. IT index k = 2 (QTM) and k = 1 (zero inflation) are interesting cases. (removed)

Counting backwards

I just wanted to link to this post by Antonio Fatas titled "Counting backwards to the next recession" so I could add this:

Monday, November 23, 2015

Is there a difference between a monetary and a barter economy?

What this means is that each individual has (up to) n-1 different consumer choice problems. Because in each of the n-1 markets, the quantity constraints are those that apply in the other n-2 markets. And that other n-2 is a different set of markets in each of the n-1 cases.

Emphasis in the original.

How many goods are there? Some estimates say that Amazon has 160 to 200 million products. Therefore each individual is solving on the order of a 100-million-dimensional consumer choice problem. That linear programming problem would take a typical modern desktop computer about 300 hours to solve.

Obviously there is some factorization happening (I don't really optimize over all goods, but probably classes of goods and narrow the focus to a specific need at a specific time), but it's still unrealistic.

Note that there are n(n-1)/2 markets in Nick's formulation in a barter economy, meaning the problem is even worse!

Since both the barter economy and the monetary economy are necessarily radically simplified relative to the full problem in both cases, does it really make sense to talk about how a monetary economy is different from a barter economy?

Does it really matter how the full linear programming solution to a 100 million dimensional problem differs from a 10 quadrillion dimensional problem when we are obviously solving only a low dimensional subset of either problem at best?

Is money really making a difference here?

I'd say no. Humans are at best solving on the order of 10 dimensional problems using heuristics. It doesn't matter if the full problem is 100 dimensional or 10 quadrillion dimensional.

Business cycle for men

Ok, this was startling:

When we worry about the business cycle, is this just a case of a men's problem being considered a national problem?

There's probably a lot more to this (this was just my instant reaction), but prima facie it is pretty startling.

The housing bubble: a personal view

Paul Krugman discusses the housing bubble and the run-up to the financial crisis in his new post and I thought I'd share a quick bit personal history.

I received my PhD in physics in 2005 (a bit more than ten years ago), but after spending the previous year travelling to various institutions to give post-doc interviews and talks, meeting other post-docs and hanging out in those cities I decided that I didn't really want to do it. At least, in the US. I was offered a few other positions -- I considered one in Tawian just for the experience (I like Wong Kar Wai movies). In the end I said to myself that it would be University of Liège working with Maxim Polyakov (e.g. here) or nothing. He didn't get as much funding as he needed and so wasn't able to hire anyone that year.

That's when I decided to look into that flow of physicists into Wall Street. In one case, I considered a job as a quant in New York. But at this point in my job search, I ran smack into the housing bubble.

I had an interview at one bank that I'll just call Acme Banking. Acme Banking's chief line of business was mortgages. And I was interviewing for what turned out to be a more senior mortgage risk analysis position than I originally thought. The first interview happened in the risk analysis department that seemed more like one of those movie newsrooms where papers were flying everywhere and everyone is running around. I was in a small room and one of the risk analysts proceeded to give me the strangest whiteboard lecture about tail risk. It wasn't like college where it's all academic. Tail risk was clearly real and something to be feared.

I was called back for a second interview and I met the person I would be working for. His huge office was high up in the Seattle skyline (one of the best views I've seen besides the Space Needle). His big idea -- and the reason he was interviewing people -- was to move beyond Acme Banking's old process of granting mortgages and create something more quantitative. It was during this interview that I realized that Acme Banking had basically granted mortgages to people based on the personal feelings and intuition of their mortgage salespeople. There was almost no quantitative analysis. This was 2005.

At the height of the housing bubble, Acme Banking suddenly thought it was a good idea to understand how much money was at risk. I surmised they had recently hired a couple people already and that was why the risk analysis department was so frantic.

Anyway, I decided against that job. And Acme Banking went under/was bought out a couple years later in the financial crisis of 2008.

Instead I took a job doing research and development in signal processing almost exactly ten years ago (on Nov 18th, I started on Dec 2nd). 

Sunday, November 22, 2015

Noah's unlearning economics

So now Noah Smith is unlearning economics; he put together a list of Econ 101 fails (quoted below with commentary from the information transfer framework). First one first:
1. If you slap some quick supply-and-demand graphs on the board, it looks like minimum wages should harm employment in the short term. But the data shows that they probably don't.
Supply and demand graphs are incorrectly used in this situation by economists (as I talk about in this post on Seattle's new 15 dollar/hour minimum wage). What is comes down to is that when they set a price floor in the Econ 101 supply and demand diagram, the "solution" determined by the demand curve and the price floor and the "solution" determined by the supply curve and the price floor are no longer a solutions of the same general equilibrium equation. If this is treated correctly, then if you set a price floor X in a single market then prices generally shift to p + X and the general equilibrium is higher (supply and demand both increase).

There is also the potential that the labor market might not be ideal (the true information equilibrium price may be higher than the realized price), so a price floor actually makes the market more ideal.

Also, with Malaney and Weinstein's gauge invariant approach (linked to here with some discussion), a constant shift shouldn't do anything (except instantaneously at the time of the shift) on its own. The constant shift can only matter relative to every other market. That means it becomes model dependent: Can you shift from minimum wage labor to machines in your specific business/industry? Or will you just have to eat higher labor costs or pass them on through as higher prices? These questions cannot be solved with Econ 101 supply and demand diagrams.

And that is the key in understanding the difference. If the price rises for labor in a market, in Econ 101, it would be interpreted as a positive demand shock or a negative supply shock -- if you didn't know about the minimum wage. That implies firms should view the situation (if they just look at the price) as if there was a shortage of workers. But there aren't fewer workers in the labor pool ... in fact, there are probably more because they'd be incentivized to enter the labor pool based on people making more money (assuming they also didn't know about the minimum wage increase).

From a rational behavior perspective, this would actually massively increase employment. Firms would think labor is a scarce commodity, but there would be lots of it. The Econ 101 picture says that firms don't interpret this price increase as scarcity, but instead just hire fewer workers.

That is to say there is a radical difference in behavior based on whether you know that the going wage was because of a minimum wage law as opposed to a supply or demand shock. Which means that somehow the price is somehow carrying information outside of a simple number. This is inconsistent with Econ 101 -- the price aggregates the effects of supply and demand shocks and doesn't categorize them as to whether they are due to regulation or part of the natural functioning of the market.

So traditional Econ 101 analysis cannot be used for minimum wages (or taxes, tariffs, subsidies, etc). It's necessarily more complicated. Info Econ 101 reasoning with supply and demand diagrams would say that it shouldn't do anything. But you should really look at it as a much more complicated problem not amenable to naive analysis.

I hope that is clear, because it's important.

Now for Noah's #2:
2. If there's any sort of limits to mobility, then simple labor demand theory says that a big influx of immigrants should depress the wages of native-born workers of comparable skill. But the data shows that in many cases, especially in the U.S., the effect is very small.
I am not sure why any economist would think of this as a short run pure supply shock in partial equilibrium.

In the information equilibrium model, a positive labor supply shock would lower the price level (i.e. lower nominal wages, but not real wages) if it was unaccompanied by nominal output increase. If NGDP was constant (a requirement for the partial equilibrium analysis), that actually means since the price level goes down RGDP increases since NGDP/P = RGDP.

However, how do you add people to the labor market without increasing output unless all of them remain unemployed? Their wages are NGDP. Their output is NGDP. It is impossible to add labor without taking into account the at least possibility of adding output, so the labor market should never be considered partial equilibrium. Therefore you can't use supply and demand diagrams. You should use general equilibrium -- something more like the Solow production function -- which says extra labor means extra output.

Of course at this point you say: but the question is about the effect on native born workers.

Ah, but since output is in information equilibrium with total nominal wages (NGDP ⇄ NW) with a constant price (at least for Info Econ 101 analysis), if output goes up then total nominal wages must go up as well. If NW goes up by X ... NW → NW + X, if the wages of one group goes down, then we must have X = Y - Z (immigrants add Y > X to NW and take Z away from the existing labor force). This becomes two competing general equilibrium effects. That may be what happens, but it's not a simple Econ 101 supply shock. The Info Econ 101 tells you something more complex must be happening (if it is happening at all).

On to #3:
3. A simple theory of labor-leisure choice predicts that welfare should make recipients work less. But a raft of new studies shows that in countries around the world, welfare programs barely reduce observable work effort.
Level of work (like hours per week) is a social norm (below), so is hard to budge. In fact, it is one of the most constant things in economics. There is no way you can understand a few percent effect like this until you understand larger effects like the price level.

And finally #4:
4. Most standard econ theory doesn't assume the existence of social norms. But experiments consistently show that social norms (or morals, broadly conceived) matter to people.
So you should assume you don't know how people make decisions at the micro level. Don't worry -- this doesn't mean you have to get rid of utility maximizing agents; instead they are emergent and therefore the behavioral implications don't apply at the micro level.

Update 11/24/2015

Noah follows up at Bloomberg View. He suggests a focus on empirics to protect against naive application of theory. I think "scope conditions" would be better. On a related note Cameron Murray has a good post up comparing economics to religion where I commented:
It pretty amazing to me how much economists think Econ 101 applies to real world problems ... as a contrast Physics 101 is almost universally considered an idealization of real world problems. An academic physics professor naively applying frictionless motion to the real world would be considered a nutcase.

I think scope conditions may be the common factor. A person that naively applies an idealized religious framework is a fundamentalist. Someone who applies that framework with an understanding of its scope conditions is a 'religious scientist'. For example, the issue of abortion was not actually addressed by Jesus or the New Testament. A fundamentalist figures out a way to make the framework address an issue beyond the scope conditions of the original theory. A 'religious scientist' would recognize that the framework doesn't address the issue.

Econ 101 seems to be taught and followed more as religious fundamentalism in this regard. The limitations (scope conditions) of Econ 101 aren't taught and it is naively applied to problems outside its scope. That is economic fundamentalism in the same sense as religious fundamentalism.

Teaching the idealized theory along with its limits seems a better idea than teaching complex empirical analysis to me.

Update 11/26/2015

I just wanted to add this graph that simply illustrates the issue with the Econ 101 supply and demand diagram explanation of the effect of a minimum wage and the Info Econ 101 explanation:

In this diagram the equilibrium price is 1 at ΔQ = 0. In the Econ 101 explanation, if you set a minimum wage of 1.6 (horizontal line), then demand falls to the point indicated by the dashed line. However, the blue demand curve isn't the solution to the differential equation with a minimum price of 1.6, but rather the solution for a minimum price of zero. You can't change the boundary condition and assume the demand curve will be the same.

Saturday, November 21, 2015

Does market monetarism exist in reaction to fiscal stimulus?

I kind of went on a rant in response to a comment where I said something that I think crystallizes something:
[Market Monetarism (MM)] seems like a theory made up solely for the purpose of justifying not doing any fiscal stimulus. It has very little use besides that purpose ...
Sounds harsh, right? But let's look at some history.
A pretty startling coincidence, right? Sumner is linked to by Tyler Cowen on Feb 25th (which causes him to blow up) because Cowen is on the lookout for any links that say fiscal stimulus doesn't work. My intuition for why Sumner blows up on the Internet is that Sumner and what becomes MM are a very good post hoc rationalization of people who read MR and are already opposed to fiscal stimulus (or maybe just Obama). Nothing goes viral like stuff that gives you a reason to hold on to your existing beliefs. In an unsurprising turn of events, Sumner later decides to leave Bentley and join Cowen at Mercatus.

As my wife would say: market monetarism exists in reaction to fiscal stimulus [1].

So the timing is right. What about the content?

Well that initial blog post from Sumner consisted of simply a declaration that monetary policy was effective.
Premise 1: The only coherent way of characterizing monetary policy as being either too”easy” or “tight” is relative to the policy stance expected to achieve the central bank’s goals. 
Premise 2: “Monetary policy can be highly effective in reviving a weak economy even if short-term interest rates are already near zero.” 
Premise 3: After mid-2008, and especially in early October, the expected growth in the price level and nominal GDP fell increasingly far below the Fed’s implicit target.
It adds in that it's not just that monetary policy is effective, but caused the recession. You have to logically include that: if the Fed can control NGDP and P, why did we have a recession? Because the Fed let expected NGDP fall ... it let the recession happen. It still controls it, just badly.

MM as a theory only addresses the price level P, nominal output NGDP (and thus real output RGDP = Y = NGDP/P), nominal wages (note that 2.1 NW ≈ NGDP to a very good approximation, so this isn't necessarily an independent measure) and unemployment U. Exchange rates and stock market are seen as indicators for expected NGDP and P, so they're not independent measures.

The musical chairs piece is just a mechanism for NGDP to affect unemployment. Wages have to be sticky, otherwise monetary policy has no effect. Thus wages are sticky is equivalent to saying the Fed controls NGDP and P (i.e. nominal changes have real effects). It is also equivalent to saying NGDP shocks are the cause of unemployment.

Expected NGDP and P are controlled by the Fed, so therefore the Fed controls the LHS of

NGDP = C + I + G + NX

And the RHS is just a "waste of time" (another early post from Feb 23rd, 2009). It's an accounting identity so you can't say G →G + dG means NGDP → NGDP + dG. Something has to fall like I or C.

That is to say MM consists of just the pieces of 1960s monetarism necessary to tell us why fiscal policy is ineffective demand management. There are no other pieces or implications [2]. That's why I said what I said in my rant. If you think MM is "new", it's really only new in its elevation of expectations. Old monetarists thought expectations had an effect, but did believe you had to actually print money sometimes. They were asymptotic "people of the concrete steppes". In fact, the reason why old fashioned monetarism didn't have Green Lantern expectations (i.e. sheer willpower on the part of the central bank can bring about price level, inflation or NGDP targets) was probably because they thought the idea unrealistic.

My rant was itself in reaction to a comment that asked why I go after Sumner so much. It's because the emperor has no clothes. He is the "head" of an economic "school" that would not exist without the liquidity trap argument for fiscal stimulus. And the reaction to the liquidity trap argument is to say: "nuh-uh". Quite literally, MM just repeats back the empirically invalid premise the liquidity trap argues against. MM simply reminds us of why macroeconomics needed the liquidity trap argument to say fiscal stimulus could work in the first place.
Economist 1: The central bank sets inflation through expectations of monetary policy. 
Economist 2: But that is inconsistent with data from Japan.  
Krugman: What if there is a liquidity trap where it becomes difficult to generate inflation with monetary policy?  
Sumner: There is no liquidity trap. The central bank sets inflation through expectations of monetary policy. 
Economist 2: Weren't you listening to the conversation? We already said that but that's inconsistent with Japan and now the US.
Sumner: Nuh-uh. It's just not a true Scotsman, I mean, inflation target. They just didn't want it enough.
Now one might say the liquidity trap argument exists in reaction to monetarism. But the modern form of it actually exists as an attempt to explain Japan's lost decade(s) ... and why monetary policy seems ineffective there. Unlike for MM, the timing is off. Monetarism had existed for decades and was basically incorporated into so-called "New Keynesianism". The liquidity trap argument exists in reaction to data.

How does MM explain Japan's lost decade(s)? By saying the BoJ is did it on purpose. There's no mystery to ineffective monetary policy. This explanation of Japan is mentioned a couple weeks after Sumner starts his blog:
... to call [the situation in Japan] a “trap” one would have to assume that the BOJ sincerely wanted to end the deflation. Unfortunately, there is no evidence (in their behavior) that they did.
This also explains the antipathy between Scott Sumner and Paul Krugman, the primary popular proponent of the modern liquidity trap argument. MM and the liquidity trap are like matter and antimatter; each is the negation of the other. The liquidity trap argument says it is too hard for the central bank to set expectations; MM says it's easy (if you really want to). At least when the natural rate of interest is negative -- Krugman and Sumner are generally in agreement when the natural rate is positive. That's because MM is basically the parts of 1960s monetarism that were not obviously inconsistent with data (like constant velocity) and were picked up by New Keynesianism.

The best response when any market monetarist argument is: Yes, we know. But that turned out to be inconsistent with the data unless you ignore central banks' public statements on the stance of monetary policy.

Which is why Sumner says NGDP determines the stance of policy, not central banks' public statements. The Fed or BoJ can say it's targeting 2% inflation, but apparently it doesn't mean it unless it shows up in NGDP.

So in summary market monetarism:
  • Starts in response to the Obama stimulus (ARRA)
  • Is spread by the right-leaning/libertarian econoblogosphere 
  • Consists of only the tenets of 1960s monetarism needed to say fiscal stimulus is ineffective
  • Is exactly what the 1990s liquidity trap argues doesn't work
  • Doesn't have any other consequences [2]
It's the immune reaction to an infection of Keynesianism that lets people keep thinking government spending is bad [3].

Although on a much smaller scale, it's analogous to the resurgence of mysticism and mysteries in the Catholic Church in reaction to the reintroduction of Greek logic to Europe. Yes, the stories and the magic tricks seem illogical, but it's a mystery unknowable by mortals! The wine is turned into blood during the Mass, but still tastes like wine because ... mystery! I wrote about this more on my old blog. But the conclusion there was that such theories, whether religious or political, do not fail when faced with logic or data.

This last piece gives me pause. In writing this blog post on my flight home from a work trip, it has dawned on me that this means MM probably wouldn't be assailable by logic because it's not really about theory or understanding the world. It's a reaction to a theory and a defense against understanding of the world.

So this will be my last post on Scott Sumner and Market Monetarism. In the future, I'll just link back here or the many other posts that take it on (in the process of which I learned a lot) in response to questions about it. I encourage its followers to look deeply into themselves and ask: What do I really want out of market monetarism? When you ask yourself this, remember that confirmation bias is a heck of a drug. If you just want comfort about the state of the world, that's fine! For some people religion has that effect. But if you're genuinely trying to figure out how macroeconomics works, you should accept the fact that macroeconomics is not well understood. No one really knows how central banks interact with the macroeconomy. 

PS I'll still argue with Nick Rowe though because that's usually about math or specific traditional macro models.


[0] There is a possibility that Sumner started his blog after Paul Krugman made his famous "dark ages" swipe at Chicago macro (a day before the ARRA passes the House). The subject is still fiscal stimulus, so that wouldn't necessarily change the main thrust of this post. 

[1] My blog exists in reaction to Noah Smith saying macroeconomics doesn't work. I started it a week after this post.

[2] Ok, so there's not zero additional implications of MM. Sticky wages mean that things designed to "artificially" (I.e. not with monetary policy) raise wages won't work. But that's really just a restatement of the thesis that fiscal policy is ineffective. Sumner adds that taxing wages is contractionary because of sticky wages, but that can't really be true because that is fiscal policy. It would stand to reason that sticky wages implies tax cuts would be expansionary (about half the ARRA was tax cuts).

But wage tax increases is contractionary fiscal policy! MM exists in reaction to fiscal stimulus! So therefore, wage taxes are contractionary, but somehow wage tax cuts aren't expansionary (because the Fed controls expected NGDP).

So these extra implications are either redundant or confused. Potentially both.

[3] MM is also the last gasp of Milton Friedman's "compromise": government bureaucrats are inept, but the central bank (run by government bureaucrats) doesn't have to be.

Monday, November 16, 2015

Scott Sumner doesn't understand other macro models

Scott Sumner (how many times have I started a post off with that?) tries to troll the left-econoblogosphere:
Suppose we were back in the 1990s, and unemployment was 5.0%. But now suppose the economy was growing slowly due to slow growth in the working age population and slow growth in productivity. A “Pop Keynesian” says that we can solve this problem with fiscal stimulus. What do the smart 1990s Keynesians say in reply? 
What do they say today?

Tyler Cowen re-trolls us all by quoting it and adding a new title: Questions that are rarely asked. I guess I'm feeding the trolls.

Scott's question is rarely asked because it's a dumb question. The US in the 1990s wasn't in a liquidity trap as inflation was 4% and interest rates were higher. Many economists didn't think a liquidity trap as an issue, although some were concerned at the time about the proper inflation target to keep away from the zero lower bound on nominal interest rates. No Keynesian would suggest fiscal stimulus in the 1990s in the US. There is a difference in the answer between 1990s and today because today the US (and Japan) seem to be persistently undershooting their inflation targets, a sign of a lack of traction for monetary policy and a liquidity trap. In a liquidity trap, fiscal stimulus becomes a viable option.

This is one of those dumb 'gotcha' questions where Sumner is saying: You say X now when you used to say Y -- ha ha! Really what is happening is that f(x, y, z, t = 1990) = Y and f(x, y, z, t =2015) = X.

And Scott would understand that if he understood the Keynesian model. But Scott Sumner does not understand the Keynesian model.

He doesn't know what defines a liquidity trap.
He doesn't know whether an independent central bank matters or not.
He doesn't know the effects of fiscal stimulus or what qualifies as fiscal contraction.
He doesn't know how to measure the impact of fiscal stimulus or contraction.

We should take his pronouncements on anything Keynesian with a grain of salt. Better yet: ignore them altogether.

And it doesn't matter if Keynesian economics, broadly construed, is wrong. Maybe it is. Scott doesn't understand it either way. It could be the aether or general relativity. It doesn't change the fact that Scott doesn't know what it is.

What Scott does is substitute market monetarism (Green Lantern Institutional Bank theory) for Keynesian economics and then tries to understand Keynesian policy prescriptions in terms of GLIB theory.

However ...

I've recently discovered that his inability to understand Keynesian economics (and instead inexplicably filling the gaps with GLIB theory) is part of a more general inability to understand any macro model besides market monetarism.

My counterfactual is that had NGDP kept growing at 5% in 2008-09, then RGDP would have also kept growing (although it would have slowed slightly for supply-side reasons) and I claim that wages would have continued growing at about 4%. An RBCer would not agree. In their view the counterfactual result would be high inflation and high nominal wage growth, indeed wages soaring at perhaps 10%/year, or something like that. And because wages would have soared by 10%, the stable 5% NGDP growth would lead to 5% fewer hours worked, and the unemployment rate would soar from 5% to 10%. RBCers don’t believe than nominal shocks have real effects. The Great Recession was caused by real factors, in their view. 
The math fits, but how plausible is that counterfactual? And keep in mind, BTW, this is the ONLY possible counterfactual to my claim that stable NGDP growth would have maintained high employment in 2008-09 ...
An RBC economists would say that if NGDP and RGDP had kept growing at roughly the same rate as it had before, then there couldn't have been any real factors causing a recession. Massive changes in wages isn't the only possible counterfactual -- nothing happening at all is also completely consistent with Scott's counterfactual path of NGDP and RGDP. RBC economists don't believe you can have a recession but keep NGDP and RGDP growing at their trend rates. They think Scott's GLIB theory counterfactual is impossible while still having a recession. I have no love for RBC, but at least I get it. 

I have a rule of thumb:
All existing theories are superficially consistent with the qualitative behavior of the data.

That is to say there is probably some way that any theory can describe some set of data or some thought experiment; your job as a critic is to find out what that is and present your critique in a way that takes that into account. It's essentially being charitable towards other theories (in a way, a corollary of  "Feynman integrity"). That's why I knew there had to be a problem with the way Scott was critiquing RBC theory -- there had to be a way to make it superficially consistent with his counterfactual.

My problem with market monetarism (GLIB theory) is that finding that plausible consistent story is too easy. Any data or thought experiment can be made to fit the theory. Market monetarism isn't falsifiable. There are no states of the world that are inconsistent with its statements. That's probably why an army has gathered behind Scott (Noah Smith makes a joke about terra cotta grad students).

Scott doesn't take my rule of thumb into account in his criticisms of Keynesian economics or any other macro theory. He can't! He doesn't understand the other theories [1]! And of course it's easy to show how something is consistent with market monetarism -- everything is! Counterintuitive macro results are easy to understand if you have a theory that can't be wrong [2].

So you have the Dunning Kruger effect colliding with an unfalsifiable theory, and the result is glib criticisms.



[1] I think this goes a long way towards explaining why he doesn't understand information equilibrium even though when he wrote down his model, he wrote down an information equilibrium model.

[2] If you think market monetarism is falsifiable, please let us know. But I can assure you that you are wrong. All you have to do is take the supposedly unobservable state X and add and expectations operator to make E[X]. If it can be expected, it can come out of market monetarism.

Sunday, November 15, 2015


Scott Sumner says:
The 2% inflation rate since 1990 would be an amazing coincidence, almost a miracle.
Is it?

This is another place where reasoning using scales and dimensional analysis would really help. A 2% inflation rate sets a time scale (it is per year) so t = 1/π. For it to be a miracle (i.e. a fine tuning problem) that the Fed's (implicit) target and the inflation rate to match up there would have to be no other scales T ≈ t. So we should ask: Are there any other time scales T in a nation such that T ≈ t  from 1990 on besides the central bank target? 

How about say ... labor force growth?

Actually employment growth (T = 1/λ) matches up pretty well too, including the times before 1990. Nearly all of the major differences are recessions:

The information equilibrium model gives a mechanism for t = 1/π ≈ 1/λ = T for this latter example through nominal shocks.

Something that would be a major coincidence (i.e. finely tuned) if it wasn't the result of some underlying theory is the Grand Unified Theory (GUT) scale:

Another place where the value of a parameter considered something of a miracle is the strong CP problem. It's somewhat of a miracle that the CP-violating term of the QCD Lagrangian is about a trillion times too small. It's such a fine-tuning problem that the axion was proposed to fix it.

But there exists at least one scale T that is approximately equal to t (i.e. π ≈ λ), so it's not necessarily a miracle that inflation is on the order of 2%.

Actually, it's more of a miracle that central banks chose to (implicitly) target 2% inflation. A larger target might have shown persistent undershooting earlier. However, the Fed never announced an explicit inflation targeting policy, and only has said 2% is consistent with its mandate more recently. Overall, the onset (1990s) and explicit target (about 2%) have a lot of wiggle room. And saying the Fed has targeted 2% PCE inflation does not explain the unbroken trend towards lower inflation since the 1980s (something that comes out of the IT model: see here or here).


PS I don't really want to link to the Insane Clown Posse, but their music video Miracles, which Brad DeLong frequently posts a screenshot from is immediately what I thought of when Scott called 2% inflation in the absence of effective central bank targeting a 'miracle'. In it, the ICP asks "#$%@ magnets; how do they work? And I don't want to talk to a scientist ..." Anyway, that's the rationale for the title.

Matthew Yglesias moves the goalposts on Abenomics

Robert Waldmann noticed that Matthew Yglesias has moved the goalposts from Abenomics generating inflation to some different indicators based on labor supply. Check his post out first.

I don't currently have access to my computer with the Japan model on it, but the latest data would be consistent with it as well as this re-programmed version of it:

The failure of Abenomics to generate even 1% inflation is striking on this graph.

Saturday, November 14, 2015


Tyler Cowen writes down his own (in response to Scott Sumner) and then links to a couple more macroeconomic frameworks (Arnold Kling and Ryan Avent). It's good that we're hearing about these now as all of these people have been writing about economics for some time. Cowen's frameworks at least get vintages (Cowen Macro Framework 2015 has hints of apple and woodsmoke).

I started this blog with a framework (now a paper) but didn't really realize that macroeconomics doesn't actually have an accepted framework until later, writing about it here for example. I guess that's not really true: apparently macroeconomic conceptual frameworks are like fingerprints ... unique to each economist.

A conceptual framework is supposed to codify your starting point for understanding a problem. It is not supposed to be a specific model (we call that a model, which in his defense, Scott does) and it should not include specific effects. Importantly, your framework should not make major assumptions about key questions you plan to address with your framework.

The Platonic ideal of a framework is quantum field theory. It is not a single model and doesn't have any specific effects. It tells you what you should write down to start tackling a problem (a Lagrangian with your proposed particle content or symmetry principles, for example). Quantum field theory doesn't make assumptions about what a grand unified theory is or what forces exist in nature at what scales.

The information transfer framework is not a single model, and it doesn't have any specific macro effects. It tells you what you should write down to start tackling a problem (an information equilibrium or information transfer relationship, see the paper). It doesn't make assumptions about what a recession is nor whether monetary policy or fiscal policy is better (in fact, both can help under different circumstances).

All of the frameworks above fail all of these key tests.

Scott Sumner's framework

Sumner's framework is probably the best one, but it assumes monetary policy controls the economy. Sumner calls it a model, and that is a perfectly valid thing in a model. It also assumes sticky wages, making it more of a model.

Tyler Cowen's framework

Cowen assumes not only sticky wages, but a specific mechanism for sticky wages (morale and unions and 'irrational' behavior after unemployment). And this mechanism leads to a cause of 'nominal' recessions. Mostly Cowen's framework however appears to be a list of things that are poorly understood (employment as a matching problem, optimal monetary policy, real interest rates) -- more a research program than a macro framework.

Arnold Kling's framework

While there is a funny bit (see the PS), Kling also assumes the cause of unemployment (incorrect specialization). Additionally, he seems think that figuring out most of the issues involved in specialization (and therefore recessions) involves non-mathematical models. Maybe recessions can't be described by math but saying they aren't is a major assumption not just about recessions but about math.

Ryan Avent's framework

Avent defines an economy with too much demand (in the limit where AD >> AS) as one with high inflation and one with too little (in the limit AD << AS) as one with unemployment and deflation. This is interesting: it basically assumes the AD-AS model is a limit of any macro theory. But again, that is more of a specific model than a framework.

Avent assumes monetary policy is 'almost always' the way to produce too much demand which is supposed to the the policy goal (or at least err on the side of too much). He says "Don't subsidise debt" which is likely due to some kind of specific effect. We should probably subsidize debt a bit because people are risk averse relative to what a rational agent would be, but that also seems like something you should figure out with a framework -- not assume from the start.

Who else?

I bet many of you are asking: what about other economists? Well Paul Romer and Dani Rodrik seem to take the nihilistic approach to frameworks: there's no general framework. 

However, Romer does use intertemporal utility maximization in his famous paperPaul Krugman is part of the traditional school of intertemporal utility maximization as well, which is a framework! Turns out it's wrong as far as what it says about microeconomics (humans don't really have consistent preferences), but being right isn't a critical property of a framework (and utility may just be emergent). This framework has lots of models: some with sticky prices, some without. It covers both John Cochrane's model and Michael Woodford's model Noah Smith links to here. Stephen Williamson and David Andolfatto have a paper that starts with this framework as well.

I'd probably split economics into these two classes: those who use the intertemporal utility maximization framework and those who are just philosophizing. And by philosophizing, I mean implicit theorizing and/or making stuff up (or as I put it better here).


Update 11/15/2015

Nick Rowe jumps in with something that is also not a conceptual framework, but rather a vague model or a collection of priors. Among other things his framework for understanding macroeconomics -- where one of the big unsolved problems is what is a recession -- includes a specific definition and mechanism of a recession.

If I said I was a doctor studying Alzheimer's and my conceptual framework included a tenet that Alzheimer's disease was defined by amyloid plaque build-up (rather than, say, the stereotypical symptom of memory loss) and lo and behold I put up some micrographs of amyloid plaque build-up in a neuron and said that caused Alzheimer's ... exactly what is my conceptual framework helping me understand?

A macroeconomic framework should not postulate what a recession is. You should use the framework to figure out what a recession is.


PS: There is one rather funny one from Arnold Kling:
3. Arriving at sustainable patterns of specialization and trade requires two types of adjustment: static adjustment and dynamic adjustment.

And there are two types of people: people who group things into two groups and people who don't (I'm part of the former as you can see from the last paragraph above). Given static and dynamic are logical complements of each other, you're pretty much assured this is true. Things are either static or dynamic. But then Kling goes and screws up a logical tautology. Seriously what is static adjustment? Is it anything like static change? He explains them immediately afterwards (in 4 and 5), but I really wouldn't have guessed the result. I thought prose-y economic writing was supposed to exist because it made things easier to understand ... 

Anyway, Kling basically says static adjustment is amenable to mathematics and dynamic adjustment isn't (I'm guessing the reason it can't be explained with math is lack of imagination by Kling; it sounds like tâtonnement). That makes dynamic adjustment a bit like the colloquial definition of Artificial Intelligence: that which hasn't been programmed yet. 

Friday, November 13, 2015

The baby-sitting co-op as an information transfer model

Paul Krugman has called the baby-sitting coop -- see here for original paper [pdf] -- his "favorite economic parable". It explains how recessions can be the result of poor monetary policy, that a lack of money (or in this case, baby-sitting scrip or coupons) can lead to lower output (fewer baby-sitting jobs). I've talked about it briefly before here. Let's put this parable into a formal information transfer model. Start with information equilibrium between the scrip (S), each coupon good for 30 minutes of babysitting, and the babysitting jobs (B).

P : B ⇄ S

Note that a babysitting job can potentially cost several coupons. That is our shorthand for a differential equation (see the paper) that has the solution (in general equilibrium):

B/B₀ = (S/S₀)ᵏ

The 'price level' is:

P = k (B₀/S₀) (S/S₀)ᵏ⁻¹

With this model, we have supply and demand curves in partial equilibrium where X stays near X* on the time scale that Y (the other process variable) changes (see the paper, with ΔX ≡ X - X₀):

Already we have some basic mechanics described in the paper and by Krugman:

  • Depending on the value of k, the price level could be stable (k = 1), rise (k > 1, inflationary pressure) or fall (k < 1,  deflation)
  • If scrip S falls, then babysitting B will fall in general equilibrium. Printing more scrip will allow babysitting to rise.

However, there is another bit of dynamics not captured in the mainstream economic model version that is present in the information transfer framework. It's called non-ideal information transfer. In general, the information in the spatial and temporal distribution of babysitting jobs is greater than the information in the spatial-temporal distribution of scrip (not everyone who wants to babysit can babysit) so:

I(B) ≥ I(S)

When these are not equal, I write this as an information transfer (not information equilibrium) relationship:

P : B → S

And the best we can do is put a bound on the non-ideal price level P*:

P* ≤ k (B₀/S₀) (S/S₀)ᵏ⁻¹

And if B* is the non-ideal level of output and P is the ideal price level, then

P* S/k = B* ≤ B = P S/k

Which means you can have a recession without it being caused by a lack of scrip. And this should actually happen sometimes. If you say there are a lot of time periods (say weeks) and a lot of people in the babysitting coop, even random behavior will tend to saturate this bound. But occasionally, it will be less even if there is sufficient scrip. It could look something like this:

The line at 100 represents information equilibrium level of output B. At time t = 150, we have B* ≤ B (a recession) due to non-ideal information transfer. Now it doesn't have to be random behavior causing it (it still could be) -- it could be a broken market or some complex behavior such as a subset of the coop deciding to go out all at the same time. Or everyone works for the government and there's a shutdown (the coop story was originally in Washington DC).

Also, even the non-ideal information transfer could be alleviated by a temporary increase in the level of scrip (which moves the information equilibrium bound up):

The babysitting coop is traditionally used as an example of how nominal shocks (to the quantity of money) can have real effects on output, and how monetary policy can be useful in setting things right. The information equilibrium picture reproduces this basic picture. However it adds that recessions may also be caused by other factors (financial panics, wars, market failure, human behavior) that have nothing to do with monetary policy. Instead of the financial crisis being caused by tight money in 2008, it might not have been the Fed's fault at all.

Basically, the information transfer framework says that things like recessions may not arise out of the behavior of rational agents (which tend to produce information equilibrium results).