Short takes

1.  Lars Christensen is now available to do public speaking.  Please hire him.

2.  Speaking of which, Warren Buffett needs to spend more time reading market monetarist blogs.  But at least he understands labor economics.

3.  Ed Dolan has the wrong test of fiscal policy.  He includes lots of eurozone countries for which monetary offset does not apply. Numerous researchers who have done the correct regression (for countries with an independent monetary policy) find no effect.

4.  David Beckworth says Fed policy contributed to the housing boom.  My view is that a counterfactual of 5% NGDP growth, level targeting, would have seen a smaller housing boom, but only slightly smaller.

5.  Marcus Nunes sent me a long NYT article on the question of whether the Fed should raise its inflation target to 4%. Space devoted to the level targeting option?  Zero.  Space devoted to the NGDP targeting option?  Zero.

6.  David Levey pointed me to another British blog advocating NGDPLT, by Anton Howes.

7.  On June 18, 2008, the Fed thought 2009 would be a pretty decent year.  At the time they knew full well that the US housing market was in a horrific free fall.  What they didn’t know is that they had set their interest rate target at a level that would soon produce the biggest fall in NGDP since the 1930s.  It began within days of this forecast, long before Lehman failed.

8.  I wish journals like the AER had more articles similar to this Scott Alexander blog post.  And this excellent follow-up is quite, well . . . depressing.

9.  Are Texas and Arizona dramatically less racist than New York and New Jersey?  Maybe.  But I wouldn’t draw that conclusion based on this study.

10.  Are you a white or black person, and feel like you are a nobody.  Go to China and get involved in real estate promotion. This very amusing 6 minute video shows how.

11.  Scientists like experiments.  Businesses likes experiments.  Artists like experiments.  Governments hate experiments. More than 200 governments control over 50 million square miles of land.  Many of the governments are pathetic, and much of the land is useless.  But they won’t allow even 5 square miles for an experiment in governance.  Not one of them.  Not even the worst of them.  Not even on worthless land.  Paul Romer doesn’t seem that scary to me.  What are they afraid of?

Yes, I know, politics isn’t about good governance.

12.  Speaking of which, lots of people (including me) like to pose as rebels, but the blogosphere contains only one true radical.  Everyone else is struggling to overcome homo sapien bias.


Tags:

 
 
 

30 Responses to “Short takes”

  1. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    3. May 2015 at 12:11

    The problem with 4% inflation is that shortly after you get 6%, and then 8%….

    Brazil has a 4.5% target. We had close to that one year out of 12. Then we moved to 5.5 to 6 a couple of years, then 6-6.5 a couple of years, now we have 8% …. and interest rates have gone up to 13.25% and the tightening cycle is not over.

    It is very costly to bring inflation back down again. NGDP targeting is much better. Sumners 4.5% or 5% NGDP is much better than 4% inflation targeting. But I would not do level targeting. If there is a lasting not so good productivity period all NGDP LT will create is higher inflation and you will only realize that three years down the road, if the target number is too large. Sumner’s 4.5% is well chosen. People Speak up, get the conservatives on board with a lower target, if you have to compromise to get support.

  2. Gravatar of Kevin Erdmann Kevin Erdmann
    3. May 2015 at 12:47

    It’s amazing how the Fed spent the summer of 2008 observing a shocking series of trends and events – collapses in home prices and inflation expectations, bank failures, rising unemployment – and explicitly taking discretionary hawkish positions in the face of it, and now they along with seemingly everyone else refer to that period as a time when monetary policy just didn’t have the potential to stop the crisis.
    Ironically, the fact that they have not been accountable to that leaves them open to the complaint that they are Wall Streets lapdog, bailing out powerful interests. So they still take blame. Surely it would be better for them to get blames for the right thing.

  3. Gravatar of marcus nunes marcus nunes
    3. May 2015 at 13:05

    I don´t think the Fed had anything to do with the housing boom, but everything to do with getting a depression!
    https://thefaintofheart.wordpress.com/2015/05/03/bernankes-amnesia-caused-the-depression/

  4. Gravatar of marcus nunes marcus nunes
    3. May 2015 at 13:08

    Contra Beckworth:
    https://thefaintofheart.wordpress.com/2015/04/25/the-2002-04-period-in-the-limelight-once-again/

  5. Gravatar of Don Geddis Don Geddis
    3. May 2015 at 13:40

    @Jose Romeu Robazzi: Level targeting is even more important than NGDP targeting. It assigns clear responsibility for monetary policy failures. The central banks can’t keep missing targets, year after year, and offer only excuses. And it provides predictable long-run value for the currency, not just short-run. Trying to estimate the true net present value for a 30-year mortgage is close to impossible, without a level target.

    If you only allowed me one of the two, either an NGDP growth target or an inflation level target, I’d choose the level target. (Of course, NGDPLT would be far better than either of those.)

  6. Gravatar of E. Harding E. Harding
    3. May 2015 at 14:51

    “Are Texas and Arizona dramatically less racist than New York and New Jersey? Maybe.”
    http://www.unz.com/isteve/more-liberals-states-have-more-racial-inequality-in-imprisonment/

  7. Gravatar of JonathanH JonathanH
    3. May 2015 at 15:07

    Scott, off topic. I’ve been thinking about the NGDP prediction market and notice that expectations of NGDP seem to fluctuate as frequently as stock prices.

    If this is true, is the NGDP prediction market as good an indicator of the health of the economy as the stock market?

  8. Gravatar of benjamin cole benjamin cole
    3. May 2015 at 15:20

    As an interim step to NGDPLT, then IT banding might not be so bad. It might be politically possible. The RBA’s IT band of 2% to 3% seems okay. I would prefer 2.5% to 3.5%.

  9. Gravatar of Econymous Econymous
    3. May 2015 at 15:46

    Here’s a fun one for you:

    http://www.terryburnham.com/2015/04/chicken-little-portfolio-suffers-in.html?m=1

    The first half of the post is about how lousy his investment advice has been, but skipping down the page to #2 shows a real doozy of a claim about inflation.

    Apparently big economic crashes just happen and monetary policy determines whether they’ll be deflationary or hyperinflationary crashes. The man who wrote this used to teach at Harvard.

  10. Gravatar of CMA CMA
    3. May 2015 at 16:03

    “My view is that a counterfactual of 5% NGDP growth, level targeting, would have seen a smaller housing boom, but only slightly smaller.”

    A 5% ngdplt achieved through drops of emoney would never have required lowering rates to the same extent we saw historically becuase drops stimulate through higher money wealth, not lowering rates. Therefore housing excess would be unlikely to form.

  11. Gravatar of benjamin cole benjamin cole
    3. May 2015 at 16:09

    JR Robazzi—Is the problem with 2% inflation that thereafter you get perma-ZLB? See Japan…Europe…USA?

  12. Gravatar of LC LC
    3. May 2015 at 17:14

    I might be paranoid, but something is off about the NYT video you linked to. The Chinese mandarin accents are just too perfect, the colloquial expressions too western and urbane. This clip is more like a Hollywood production than a documentary. (Note I don’t doubt some of the events depicted are actually going on in China, but something just seems too artificial and too produced for real world documentary.) I tried a search for the director and some of the people and firms in Google and Baidu, and results (even in Chinese) linked back to the video and not much of anything else.

  13. Gravatar of ssumner ssumner
    3. May 2015 at 17:36

    E. Harding, As I said, maybe. But that’s not the test I’d use either. Of course there are lots of ways of defining racism, so I actually wouldn’t have a lot of confidence in any single test.

    Jonathan, In my view the Hypermind market is not very efficient yet. I’d guess the liquidity is too low at this point. It’s a work in progress. When it becomes efficient it will be more stable.

    CMA, I have no idea what “drops of emoney” means, but it doesn’t sound promising.

    LC, It’s too good not to be true. 🙂

  14. Gravatar of Ray Lopez Ray Lopez
    3. May 2015 at 19:58

    It is surprising and amusing to see Sumner supporter Robazzi concerned about inflation running away if you set a higher inflation target. After all, if Sumner’s NGDPLT is correct, you can always cut back on inflation as if by pulling on a string by decreasing the money supply, simple eh? But we all know it’s not so simple and Robazzi is correct. However, Robazzi exaggerates inflation’s pernicious effects, as Brazil for forty years had hyperinflation and the drag on GDP was only 3-5% (negative) a year over that period (source: Calomiris’ latest banking book), which is not ‘that bad’. The real problem is the total govt debt. If it becomes big enough then government will try and hyperinflate it to zero, as was done in Germany (that is, the -3%/yr must increase to say -35% a year for outstanding debt to be halved every two years, which is easier said than done as people adjust quickly to hyper-inflation).

    Further, as I posed to the board and nobody answered a while ago: at some point in a recovery from a recession NGDPLT must coincidentally have the same affect on the economy as higher inflation targeting, so why not adopt a higher inflation target? (BTW I am against any target as I believe money is largely neutral). I am waiting to hear why you people don’t believe this. Unless you believe in metaphysics such as “expectations would not be set by higher inflation targeting, but they would be set by NGDPLT (despite the fact the latter is harder for the public to understand)”. You can see then that NGDPLT is like trying to balance a plate on a pin: lots of unstable assumptions are built into the model that even true believers like you people cannot agree upon.

  15. Gravatar of Morgan Warstler Morgan Warstler
    4. May 2015 at 04:38

    Don, that’s one of my complaints with Scott’s presentation.

    Instead of NGDP vs price level versus wage etc etc.

    I think he’d get further if he just banged gong on LT – no matter what it is, set that LT and here’s the formula for making it up on a set 1 month or 2 month curve.

    Because deep down, I think Scott would take anything on an LT first. And I think the ask of LT is easier to get thru into the brains of economists.

  16. Gravatar of ssumner ssumner
    4. May 2015 at 05:59

    Ray, The claim that the public understands inflation targeting (even a tiny bit) just might be the funniest thing you ever said.

  17. Gravatar of Jim Glass Jim Glass
    4. May 2015 at 06:10

    Ed Dolan has the wrong test of fiscal policy…

    From a statistician’s point of view, here is a nice critique of this ‘spending –> growth across the OECD’ chart and claim as originally presented by Krugman…

    http://justthesocialfacts.blogspot.fr/2015/01/less-than-meets-eye.html

    Krugman says, “you can, if you like, try to argue that this relationship is spurious, maybe not causal.” … But thinking about the possibility of spurious correlation isn’t a matter of liking — it should be pretty much automatic.

    Indeed. And again, those of us old enough to remember the pre-politicized, Nobel-earning, pre-2000 Krugman, can’t help but think that *that* PK knew full well that competent statistical analysis considers the possibility of spurious correlations and causation issues as a first-thing up front before making a presentation — not as an after-thought, “if you like”. And that that PK, with his typical tact, would have roasted such an “after-thoughter”.

    Another example of the toxic result when the worm of politicization eats it way into even (especially?) the smartest mind.

  18. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    4. May 2015 at 06:16

    @Ray
    I don’t like the inflation tax, but that is not the worst problem inflation causes. I think that inflation creates coordination problems in the economy, and trashes a country’s industrial structure. Brazil today produces absolutely nothing that other countries want to buy, that was not the case even 20 years ago. We do zero research on the technological frontier, because everyone who put money on the capital goods sector was killed by the macro instability that inflation is either a sympton of or a cause to.

    @Cole
    I am with Sumner here, QE can always deal with AD shortfall even at the ZLB, no need to have higher rates, in my view.

    @Don Geddis
    I get the LT commitment issue, and I think it is important, but even Sumner here has defended a lower target if LT is adopted. I can only assume that the reason behind the lower target is that if you do get a persistent negative productivity shock, the monetary authority runs the risk of only producing higher inflation … and creating instability, which is exactly we are trying to fix by replacing IT, isn’t it ?

  19. Gravatar of Daniel Daniel
    4. May 2015 at 06:34

    Jose

    I think that inflation creates coordination problems in the economy, and trashes a country’s industrial structure.

    But how can that be, since money is neutral and super-neutral ?

    QE can always deal with AD shortfall even at the ZLB, no need to have higher rates

    http://en.wikipedia.org/wiki/Fisher_equation

    You are a very ignorant man, yet you have very strong opinions on subjects you clearly do not understand.

  20. Gravatar of Don Geddis Don Geddis
    4. May 2015 at 07:14

    @Jose Romeu Robazzi: “runs the risk of only producing higher inflation … and creating instability

    Not really. Sumner has argued that macro economic damage comes from NGDP instability, not from inflation instability. It’s hard to distinguish historically, because in the typical case both NGDP and inflation are volatile at the same time. But if you had a central bank successfully implementing NGDPLT, and you had your case of a persistent negative productivity shock, the question is whether the resulting unexpected higher inflation — but in the presence of stable NGDP — would actually be a “problem” needing to be “solved”. Sumner would say no.

  21. Gravatar of Willy2 Willy2
    4. May 2015 at 11:16

    – Foreign policy (Wars in Iraq & Afghanistan) of the Bush administration helped to increased the US Current Account Deficit and that helped to inflate the housing bubble even more.

  22. Gravatar of Bob Murphy Bob Murphy
    4. May 2015 at 19:39

    Scott is this really your position?

    What they didn’t know is that they had set their interest rate target at a level that would soon produce the biggest fall in NGDP since the 1930s.

    In your view, it’s not that particular interest rate targets map to particular economic outcomes, right? You’re making it sound in the quote above that if only the Fed had cut interest rates earlier, then there would be no recession. But haven’t you spent years telling us that that is the exact wrong way to think about the Fed?

  23. Gravatar of ssumner ssumner
    5. May 2015 at 06:10

    Willy2, I doubt the wars impacted the deficit significantly. The housing boom is more plausible.

    Bob, You said:

    “In your view, it’s not that particular interest rate targets map to particular economic outcomes, right?”

    Yes, that’s right. In this case the interest rate target plus their other policies such as “communication” led to the disaster.

  24. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    8. May 2015 at 14:39

    @Daniel,
    Thanks for letting me know I am a very ignorant man. Gee, who is this Fisher guy, never heard about him before ….

  25. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    8. May 2015 at 15:02

    @Don Geddis
    I think your reasoning is acceptable for the US, where real GDP probably has low volatility. As an example to help us imagine possible outcomes, If the monetary authority does not meet its target of NGDP, let’s say 4,5% n a LT regime for two years, it gets 1.5% growth and 2.0% inflation, and it tries again in the third year, and succeeds, then there will be a year of 1.5% growth and 5% inflation, that is not very bad, although 5% inflation is hard to swallow for the US. But suppose the same situation in an emerging country, with 7% NGDP and 2.5% potential RGDP, first two years at 1% RGDP and 4% inflation, in the third year you’d get 10% inflation, and that is not very good, in my view …

  26. Gravatar of Don Geddis Don Geddis
    8. May 2015 at 17:18

    @Jose Romeu Robazzi: But your example is explicitly one with unstable NGDP growth! We’ve already agreed that unstable NGDP can cause economic damage. You can’t distinguish between NGDP harm and inflation harm, by positing a hypothetical where both are unstable.

    It seems to me that you now have a different complaint. You don’t seem to believe that a central bank running NGDPLT can actually meet its targets. Even for your “US” case, you imagine targeting 4.5% but only achieving 3.5% the first year; then targeting (more than) 5.5% but again only hitting 3.5% in the second year. That’s eight quarters of disastrous and incompetent central bank performance.

    Sure, it doesn’t matter what policy rule you give to a central bank, if they refuse to actually follow it. Or do you think that even a well-intentioned central bank somehow lacks the power to achieve a nominal target?

    I basically don’t find your examples plausible. (Also, why would an emerging country with only 2.5% potential RGDP, choose an NGDP target of 7%? That seems like a very poor choice also, again leading you to describe an implausible scenario.)

  27. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    11. May 2015 at 04:40

    @Don
    Ok, The examples are a stretch. But they are there to make a point. Even if you believe the central bank will meet its target, there certainly will be times when they won’t. I don’t think it is impossible to miss a target two year in a roll. That is why i don’t like LT. That is why also I don’t like high targets. But if you think that 7% NGDP is too high for an emerging market country, I think that there are people that comment on this blog that would go with 6% for the US without blinking …

  28. Gravatar of Don Geddis Don Geddis
    11. May 2015 at 10:10

    @Jose Romeu Robazzi: “I don’t think it is impossible to miss a target two year in a roll.” Sure, maybe by a little bit, and with random error. But your example had a second year target of 5.5%, and they only achieved 3.5% — that’s a HUGE miss, especially given that they ALSO underachieved in the immediately preceeding year (4.5% target, 3.5% achieved). That’s bordering on incompetence. Especially given that they’re getting, at the least, quarterly feedback on NGDP, not just annual feedback. (And probably, if they actually adopted NGDPLT, they’d focus on better data gathering, and might have monthly NGDP reports, or even a real-time NGDP prediction market.)

    if you think that 7% NGDP is too high for an emerging market country“. No, I don’t — but you made the additional unrealistic assumption that the emerging country had “2.5% potential RGDP“. Actual real-world emerging countries generally don’t have a 2.5% cap on RGDP growth, so questions about real-world NGDP targets don’t rely on your hypothetical examples.

    6% for the US without blinking” The US had 3-5% annual inflation during the Great Moderation in the 80’s and 90’s, with no apparent ill effects. An NGDPLT target like 5% would be likely to show VERY stable long-run inflation at around 2%. What is the danger you fear? Did you see something wrong during the Great Moderation that you’re trying to fix?

    I don’t think there’s any benefit to a US (NGDPLT) target of 6% instead of 5%, but on the other hand there doesn’t seem to be much harm either. Yet you seem really worried, and I don’t know why.

  29. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    12. May 2015 at 04:08

    I fear exactly what we are having now. If the US had a 5% NGDP targeting, after having missed it for, let’s say, 4 years now, “playing catch up” (LT) might create a year with inflation, let’s say, 1.5% above the “reasonable” 2-2.5%, if potential GDP was really on the lower range of current estimates. And then, in the following year, the monetary authority need to start tightening very hard, because inflation has “memory”, it is hard to bring it down once it sets in. On the great moderation, I don’t think it was that huge success that most people claim. First, if it was so good, it would not have been hard to keep it. Second, there was growth in inequality, and the lower middle class saw no growth in real income. I am very skeptical that 3-5% inflation should be rated “good” in any measure that takes inequality into account.

  30. Gravatar of Daniel Daniel
    12. May 2015 at 05:06

    inflation has “memory”, it is hard to bring it down once it sets in.

    So inflation is like homeopathy ?

    I am very skeptical that 3-5% inflation should be rated “good” in any measure that takes inequality into account.

    http://en.wikipedia.org/wiki/Moving_the_goalposts

    You’re approaching Ray Lopez territory.

Leave a Reply