Archive for July 2015


The NSA is going to love this

I’ve been telling people that 1984 is coming, but no one seems to care.  Here’s one more indication:

Samsung SmartTV

The Samsung SmartTV has a built-in microphone that is equipped with voice recognition technology that allows users to give verbal commands to the TV. In order for Samsung to convert your speech to text, the voice commands are sent over the Internet to a third-party for interpretation.

However, since the TV is “always on,” the microphone is recording every word you’re saying at all times. Even in its SmartTV privacy policy, Samsung acknowledges that all spoken words, including personal or other sensitive information, are sent unencrypted to the third party.

Eh, what could go wrong?  It’s not like high tech firms would give in to government pressure to invade the privacy of tech users.  No need to fear the NSA ever getting any of that info.

I’ve argued that it’s up to the younger generation to figure out how much privacy they want.  My only request is that we stop having students read 1984 in high school.  If that’s the world we want, then let’s stop pretending it’s some sort of dystopia.

Fortunately we don’t have a war that seems to go on forever, and that the government uses as an excuse to have all sorts of extra powers.

Update:  Et tu, Ford?

The wrong question

The Neo-Fisherian debate continues, and continues to miss the point.  The debate is framed in terms of whether a higher interest rate causes higher inflation.  But that’s not even a question.  Or at least it’s meaningless unless you explain whether the higher interest rate is produced by an expansionary monetary policy or a contractionary monetary policy. Central banks have the tools to do it either way.  On the other hand I am increasingly getting the impression that the New Keynesian model is incapable of handling that distinction.  Here’s John Cochrane responding to a recent Woodford talk on the issue:

This is a particularly important voice, as it seemed to me that standard New-Keynesian models produce the new-Fisherian result. i = r + Epi is a steady state in all models. In old-Keynesian models, it was an unstable steady state, so an interest rate peg leads to explosive inflation or deflation. But in new-Keynesian models, an interest rate peg is the stable/indeterminate case. There are too many equilibria, but if you raise interest rates, inflation always ends up rising to meet the higher interest rate.

What I can glean from the slides is that Garcia Schmidt and Woodford agree: Yes, this is what happens in rational expectations or perfect foresight versions of the new-Keynesian model. But if you add learning mechanisms, it goes away.

My first reaction is relief — if Woodford says it is a prediction of the standard perfect foresight / rational expectations version, that means I didn’t screw up somewhere. And if one has to resort to learning and non-rational expectations to get rid of a result, the battle is half won.

But that’s only preliminary relief. Schmidt and Woodford promise a paper soon, which will undoubtedly be well crafted and challenging.

If that’s true, then the NK model is obviously very, very flawed.  Noah Smith seems to agree that rational expectations is the key assumption:

The question of whether interest rates affect inflation in a Woodfordian way or a Neo-Fisherian way depends on whether people’s expectations are infinitely rational. Woodford’s new idea – which will certainly be a working paper soon – is that people don’t adjust their expectations to infinite order. He essentially puts bounded rationality into macro. He posits a rule by which expectations converge to rational expectations.

I have one small quibble here.  When Smith writes:

The question of whether interest rates affect inflation in a Woodfordian way or a Neo-Fisherian way depends on whether people’s expectations are infinitely rational.

He seems to imply that he is discussing the real world.  Like it would actually matter whether people had ratex. My hunch is that you can easily get either result with or without ratex, if you don’t restrict yourself to the NK model.  The liquidity effect from easy money should be able to be derived with simple sticky prices, even with ratex.  I’d rather Smith had said:

The question of whether interest rates affect inflation in a Woodfordian way or a Neo-Fisherian way in the NK model depends on whether people’s expectations are infinitely rational.

BTW, in this post I showed how you could get a Neo-Fisherian result.  That doesn’t mean I think they are “right”, just the opposite.  But I am increasingly confident that they have stumbled on something important, a serious flaw in the NK model. I’d rather people continue to assume rational expectations, and fix the model in some other way—like defining monetary policy in terms of something other than interest rates.  Stop assuming that “the central bank raises interest rates” is a meaningful statement.  It isn’t.

PS.  I wrote this a couple days ago but wasn’t sure if I was missing something, so I didn’t post until today.  Nick Rowe’s new post convinced me that I’m not missing something obvious.

HT:  Tyler Cowen



Targeting inflation and offsetting fiscal austerity are the exact same thing

Stephen Williamson has a very good post on the Canadian austerity of the 1990s. But in the comment section I think he misunderstands the concept of “fiscal offset.” First an anonymous commenter says:

Canada has offset the contractionary effects of austerity via monetary policy…

And Williamson responds:

That’s part of the point. It doesn’t look like they did. As you say, Canada has an independent monetary policy, but, post-1991 they appear to be behaving as by-the-book inflation targeters. Basically, they make an agreement with the fiscal authority about their policy rule, and then they stick to it, independent of what the fiscal authority is up to.

Sticking to your target regardless of what the fiscal authority does is exactly what fiscal offset means.  The idea is simple. A fiscal contraction would normally depress AD, causing inflation to slow.  The central bank must do enough stimulus to offset that potential decline in AD, in order to prevent inflation from falling.  If you observe the inflation rate always being on target, then the central bank is successfully offsetting any fiscal action that would have otherwise moved AD and inflation.  As an analogy, if the temperature in your house is always 22 degrees (centigrade), then the thermostat/furnace is doing an excellent job of offsetting the warm and cold fronts that move through your town.

So why do I call the post “excellent”? Start with the fact that Williamson recognizes that fiscal austerity in Canada was not contractionary.  Even better he recognizes something that all too few bloggers understand:

But it might be more appropriate to think about monetary policy in terms of the ultimate goals of the central bank.


PS.  Yes, fiscal policy has other channels besides AD, so real GDP could still change. But the AD channel is what Keynesians obsess over.

HT:  Tom Brown

Lars Christensen on the euro disaster

Lars has a great new post on the euro disaster:

The graph below shows the growth performance for these two groups of European countries in the period from 2007 (the year prior to the crisis hit) to 2015.

Screen Shot 2015-07-15 at 11.17.43 AM

The difference is striking – among the 21 euro countries (including the two euro peggers) nearly half (10) of the countries today have lower real GDP levels than in 2007, while all of the floaters today have higher real GDP levels than in 2007.

Even Iceland, which had a major banking collapse in 2008 and the always politically dysfunctionally and highly indebted Hungary (both with floating exchange rates) have outgrown the majority of euro countries (and euro peggers).

In fact these two countries – the two slowest growing floaters – have outgrown the Netherlands, Denmark and Finland – countries which are always seen as examples of reform-oriented countries with über prudent policies and strong external balances and healthy public finances.

When some of the best managed countries in the eurozone can’t even outgrow Iceland, you know that something is very, very wrong.

The labor market is (very gradually) healing

Tyler Cowen has a post on the labor market, and also links to a response by Adam Ozimek.  I mostly agree with Adam, but would like to make the points in a slightly different way.  Here’s how Tyler starts out:

One striking feature of the Thursday labor market report was the mix of declining unemployment “” now down to 5.3 percent “” and continuing sluggish wages, not to mention low rates of price inflation; read Neil Irwin. Normally we would expect all the demands for those new hires to boost wages more.  What is going on?

My initial response is “All what demands for those new hires?”  I don’t see any sign of increased demand for labor, at least in the intro provided to his post.  Instead, I see an increase in the equilibrium level of employment.  And also sluggish wages.  I can think of two possible explanations, neither of which involve more demand for labor:

1.  The supply of labor shifted right, holding down wages and increasing employment.  That’s the equilibrium model.

2.  The supply and demand for labor are stable, but actual wage rates are above equilibrium due to sticky wages.  As wages gradually adjust, the horizontal sticky wage line moves downward (relative to 4% NGDP growth) and this increases quantity demanded of labor and lowers quantity supplied. Unemployment falls.  But again, no increase in the demand (curve) for labor.

So the answer to the “What is going on?” question is, “No, normally we do not expect rapid growth in employment to be associated with fast growing wages. Normally we expect fast growth in employment to occur in depressed economies, and to be associated with slow wage growth.”

I sometimes hear it argued that there is a good news aspect to this development, suggesting that the absence of wage pressures indicates there are many more people to be hired.  I wonder if this argument makes sense.  If we don’t observe a worker willing to take a job for current wages, how is that a cause for optimism about future reservation wages for those same workers?  I would think it implies some slight pessimism about whether those individuals will end up working again.  I’m not sure those workers are going to be worth so much more in the market anytime soon.

Maybe the workers are willing to take the jobs at current wage rates.  But they can’t find them due to sticky wages.  Yes, my explanation is ad hoc, but doesn’t it fit the facts?  My explanation predicts slow wage growth.  Check.  My explanation predicts total employment growing each month at a rate much faster than the working age population.  Check.  Any other explanations fit the facts that well?

And my explanation implies optimism.  If employment is growing fast BECAUSE wages are rising at a slow rate, then the longer wages rise at a slow rate, the longer employment can rise quickly. (This is Ozimek’s basic argument.)

Just to jog your memory, the data do not indicate much of a stable Phillips curve.

That’s right, the Phillips Curve model is flawed because it assumes a stable supply of labor.  And the evidence suggests that the supply of labor is increasing.

Alternatively, you might think the employment of those remaining unemployed workers is constrained by demand side forces.  I find that unlikely at this late date in the recovery but even so, this demand side hypothesis also gives no particular reason for optimism, given a very conservative Fed.

Studies suggest that downward wage flexibility is much more difficult at the zero wage rate increase bound, perhaps due to money illusion.  So there still could be a bit of residual effects from the recession, and from the unusually low rate of NGDP growth during the recovery (which itself is a sort of shock, if workers expected this recovery to be like other recoveries.) And recall that unemployment is down to 5.3%, so the vast majority of the labor market healing has already occurred. There is room for optimism because each month we get over 200,000 new jobs, which is much faster than growth in the working age population.  If wages were suddenly rising fast, we would know that we’d be close to full employment.  But that hasn’t happened yet.

Liquidity trap models do not explain why the rate of price inflation continues to be pretty much where the Fed wants it to be, and thus they also do not explain this constellation of market forces.  There is too much labor market recovery going on.

Yup, (simple) liquidity trap models don’t explain much of anything, including the inflation rate.  They don’t account for QE and other unconventional forms of monetary stimulus.

I find the most plausible explanation to be a version of The Great Reset.  A lot of workers have been revalued by the market downwards, but most incumbents are not taking pay cuts in real terms because they have insider power.  New hires, however, are not granted equally favorable terms.  If wages are steady as new hires pick up, this is in fact upward pressure relative to the counterfactual that otherwise those wages would be falling.  Flat wage are indeed what “things heated up” looks like, and it’s a good thing we had that gas price decline to bump real wages up just a bit.

Tyler’s right about the Great Reset, and this needs to be combined with the standard AD model.  It’s one reason why the recovery has taken so long; we need even more downward wage flexibility than would be the case if the labor market were not gradually moving against the bottom half of the wage distribution.  But the last sentence is probably wrong; real wages don’t matter, W/NGDP matters.  The oil price decline did not boost NGDP, and that’s one reason why MMs correctly predicted it would not speed the labor market recovery, while conventional economists incorrectly predicted it would.