Listening to the markets (plus no more zero bound)
No one likes to be contradicted. But I must admit that the markets are increasingly pointing to the likelihood that I’ve been too pessimistic about both the prospects for growth, and the likelihood that ultra-low interest rates would persist for longer than most people assumed. What do they see that I don’t?
We know that the recent spike in US 10 year bond yields occurred after the strong jobs report this morning. But the upside surprise seemed rather modest, and the headline unemployment rate (which figures into the Evans Rule) stayed at 7.6%. Also note that the recent move toward easier (or less tight) policy in Japan, Britain and the eurozone would tend to raise global growth, putting upward pressure on global real yields.
One possibility is that the jobs data is somehow flawed. In my post this morning there was a story about a dealership in Denver that couldn’t find sales people. I tend to discount those anecdotal stories, preferring aggregate data. But there is always the possibility that lots of people who are on disability/UI/food stamps, etc., are actually working in the underground economy. Some economists claim the retail sales data are too strong relative to the official unemployment data. On the other hand, that explanation would create an even bigger mystery, why has RGDP lagged the jobs data? I suppose the argument would be that RGDP is also understating actual growth, as lots of goods and services are not being measured by the government.
Note that changes in the underground economy tend to occur fairly gradually over time, so if this is an issue (and I still have my doubts), it’s a far bigger problem for monetary policy regimes like the Taylor Rule, which depend on accurate measures of the output gap, than a policy like NGDPLT, which doesn’t require any estimate of the output gap.
Also note that nominal hourly wage growth was also quite strong in June (suggesting a tightening labor market), but on the other hand that’s only one month.
EMH fans like myself can’t get too depressed about being wrong in market forecasts, as we are ALWAYS WRONG when any asset price changes significantly in a short period of time. That’s because in most asset markets the current price is quite close to the expected future price 6 or 12 months forward. All big changes are unexpected. I get a little more uneasy, however, when I can’t quite see the reason for the change, even after it has occurred. I’d guess that in a few years this will all become clear.
My best guess is that two things have happened in recent months:
1. Growth is stronger than the markets (or I) expected.
2. The Fed will respond to any given level of growth with more aggressive “tapering” than expected.
That one-two punch seems to have driven 5 and 10 year bond yields much higher than I expected.
I use the term “tapering” because I don’t mean to suggest than monetary policy will gradually become more contractionary, indeed I expect just the opposite. But it will look more contractionary to most people.
Also note that the idea we are in a liquidity trap, which never made any sense to me, has now become completely laughable by any standard. If 5 and 10 year bond yields recently rose by 100 basis points, then there’s clearly that much room for the “liquidity effect” to reduce them again. If people keep talking about a liquidity trap, then you should just laugh in their face. (Not really; one should always be polite—but at least chuckle to yourself.)
Update: Tyler Cowen just added the following good news on Twitter:
With the liquidity trap now over, dogs no longer impregnate cats, whew!,
PS. Just to be clear, I’m not claiming that the Fed could definitely cut 10 year bond yields by 100 basis points, I don’t know that. What I do know is that if they failed it would obviously be because of the income and inflation effects, which of course means there’d be no liquidity trap. So let’s PLEASE stop talking about the “zero bound.”
PS. Mark Sadowski really needs to get his own blog. Here’s an excerpt from a great comment:
Assuming the major model type estimates are correct, then in the absence of the tax increase and the sequestration 2.144 million jobs would have been created which is over 50% more than in any other two quarter period this recovery and over double what was created in the previous two quarters. It would also have been the most jobs created in any two quarter period since the 2.251 million created in 1984Q1/1984Q2.
Tags:
5. July 2013 at 09:58
“I suppose the argument would be that RGDP is also understating actual growth, as lots of goods and services are not being measured by the government”
If RGDP is being understated than I assume NGDP is too. How big of a problem is it for NGDPT that it may not be possible to accurately measure the thing that is being targeted ?
5. July 2013 at 10:19
My explanation is that we are at a point in the recovery where a little bit of good news becomes a sunspot for an accelerating recovery. (Strictly speaking, not a sunspot, because even a little bit of good news is still actual good news, but I think its role in coordinating expectations is more important at this point than its direct economic impact.) Without a commitment to an NGDP or similar target by the Fed, the economy kind of has to set its own target, and NGDP will tend to go wherever people expect it to go, but at some point, I would suggest, there is a nonlinear response, as everyone has to get on board to avoid missing the train. I’ve set forth this idea (in more Keynesian terms, but I think I could translate it into monetarist-ese if I tried), as well as the possibility that the Fed’s taper talk may have accelerated the process, in my latest blog post.
5. July 2013 at 10:34
Rob, That’s the issue I was trying to address in my comparison with the Taylor Rule. Obviously it would be something of a problem, but much less than with the Taylor rule, or other plausible policies.
Andy. That’s possible, so let me see if this is where you are going:
1. Under an unconventional monetary regime, fiscal stimulus is relatively ineffective (albeit not entirely), as it is mostly offset by more or less QE, Evans rule, etc.
2. Under conventional policy regime, fiscal stimulus is fully offset by interest rate changes.
3. However, the Fed is willing to do more via conventional measures such as interest rate cuts than unconventional measures such as QE. Hence a growth spurt that makes a shift from unconventional to conventional more likely, will also cause monetary policy to be expected to shift from unconventional to conventional, which will reinforce the growth spurt. Is that the MM version of your claim?
If so, sounds reasonable to me.
5. July 2013 at 10:36
Isn’t there an implicit output gap in any given NGDPLT? E.g., China would have a higher ideal NGDPLT than the US because its growth rate should be higher?
5. July 2013 at 10:37
“Liquidity trap” just refers to a situation in which the Federal Reserve is unable to lower the nominal interest rate. We presume that the Fed can always lower it to 0%, so we often use the term “zero-lower-bound” interchangeably with “liquidity trap.” But look up any New Kenesian model and you will see that liquidity trap analysis applies whenever the policy-targeted interest rate is down-ward constrained, whether the floor is 0% or 1% or whatever.
In this case, short-term interest rates are still hard against 0%, but longer-term rates are rising a bit. One possible reason for this is that the Fed cannot credibly commit to keeping future short-term nominal interest rates at 0%. That is, there is an effective floor on long-term interest rates that is strictly greater than zero, and hence we may not be at the “zero-lower-bound” (depending what interest rate you choose to care about), even though we are still in a liquidity trap.
5. July 2013 at 10:41
Andy, Very interesting post, but a couple comments:
1. Is it consistent with stock prices falling on the Bernanke tapering talk?
2. I don’t like the camel analogy, it’s not intuitive. (A depression should be a broken back.)
In the Great Depression there were a few occasions when a bunch of good news appeared at roughly the same time (or bad news) triggering a sharp turn in the cycle. A negative example was October 1929, and a positive example was June 1938. In both cases the total effect seemed greater than the sum of the parts, which is how I take your claim.
5. July 2013 at 10:43
Ryan, No, inflation doesn’t matter, only NGDP/person growth matters (as a first approximation–I can think of exceptions–which don’t apply to the US.).
5. July 2013 at 10:45
Scott
What´s different this time is that inflation expectations, which had been falling for several months have taken an upward trend over the last few days.
5 year Expected:
June 24: 1.62%
today: 1.94%
5. July 2013 at 11:01
Scott, you write:
EMH fans like myself can’t get too depressed about being wrong in market forecasts, as we are ALWAYS WRONG when any asset price changes significantly in a short period of time. That’s because in most asset markets the current price is quite close to the expected future price 6 or 12 months forward.
—
Actually, you’re too hard on yourself. If an asset price changes significantly in a short period of time, that doesn’t mean you were wrong ex ante, as long as the VOLATILITY was priced into the asset.
Example: there could be an event tomorrow that informs us whether we are in state 1 or state 2 of the world. The odds, ex ante, could be 50/50, but it’s a KNOWN unknown. In that case, you are guaranteed to have a large short term price movement, though you don’t know which direction. The market is adequately pricing this (cost of hedging is high, for example)
If there’s a short term change and the price of insuring volatility is low, however, then the market has issues – classify as unknown unknowns.
5. July 2013 at 11:03
Marcus, Good point.
Matthew, Stop thinking in terms of Keynesian models and start focusing on reality. You said:
“In this case, short-term interest rates are still hard against 0%, but longer-term rates are rising a bit. One possible reason for this is that the Fed cannot credibly commit to keeping future short-term nominal interest rates at 0%. That is, there is an effective floor on long-term interest rates that is strictly greater than zero, and hence we may not be at the “zero-lower-bound” (depending what interest rate you choose to care about), even though we are still in a liquidity trap.”
What has the Fed been saying over the past few weeks? And how have the markets reacted to those promises of tighter money? Credibility is the least of the Fed’s problems right now. I doubt that even Krugman would now deny that the Fed could lower 5 and 10 year bond yields if it wished to.
If the Fed failed to lower long term yields with more QE, it would be due to an income/inflation effect, not lack of policy credibility.
5. July 2013 at 11:05
Statsguy, Good point, but I actually wasn’t trying to be very hard on myself.
I’m a sensitive soul, and don’t like self criticism. 🙂
5. July 2013 at 11:35
Two-year bond yields have not increased significantly. Thus, the market expects the Fed to keep short-term rates down for a while.
Combined with TIPS-forecasted inflation going up, it looks like the markets are expecting that zero short-term rates for ~2 years, plus decent growth, will bring about 2 % annual inflation over five years (instead of ~1.5% inflation they were predicting last month).
Zero rates and rising (albeit tame) inflation. Indeed, that is not a liquidity trap (but still not a lot of traction for “standard” monetary policy).
5. July 2013 at 11:53
Scott,
Yes, I think you have fairly well translated my idea into market monetarist-ese.
Regarding the stock market’s reaction, I’m not sure. If you look at the whole day’s action that day and compare the TIPS market with the stock market, I think they still imply higher expected real cash flows (if you assume a constant equity premium). But I don’t necessarily think we can expect markets to figure things out in one day. I guess I believe (provisionally) in sort of a slow EMH, where the smart money scratches its head for a while when confronted with new information whose potential implications it hadn’t previously considered. (This implies you can beat the market if you think faster and and are right — but people who try to think things through too quickly often come to the wrong conclusions.)
You’re right that the camel analogy isn’t intuitive. (It’s not so much an analogy, really, as a cliché forced against its will to do extra duty — which I hoped would make it amusing, but it didn’t work out that way.) For now I’m just glad I managed to make it though a whole blog post. If I have occasion to rewrite it, maybe I’ll try to think of something better.
5. July 2013 at 15:51
Ciceron, It seems to me that markets now expect a rate increase before 2 years go by. And there’s plenty of traction by having the Fed buy 5 and 10 year bonds.
Andy, Yes, I was thinking about the cash flow angle a few days ago–I probably need to rethink that.
Again, where’s that NGDP prediction market?
6. July 2013 at 10:11
Prof. Sumner,
Why don’t you buy Lars Christensen’s argument that 10-yr bond yields have skyrocketed mainly due to tight money / high demand for cash in China?
6. July 2013 at 10:11
Excuse me, tightER money in China.
6. July 2013 at 10:48
[…] When it comes to the “zero lower bound,” what’s the relevant maturity? Scott Sumner and Tyler Cowen are celebrating the end of the liquidity trap, but their argument makes no sense to […]
6. July 2013 at 16:18
Travis, I don’t see a convincing argument. The Chinese would have to buy a lot of T-bonds—-does anyone have data showing that?
6. July 2013 at 16:58
[…] See full story on themoneyillusion.com […]
7. July 2013 at 07:46
[…] When it comes to the “zero lower bound,” what’s the relevant maturity? Scott Sumner and Tyler Cowen are celebrating the end of the liquidity trap, but their argument makes no sense to […]
8. July 2013 at 12:35
Your occamist desire to exclude credit growth from your model is admirable, but leaves a lot out. If you look at the credit aggregates in addition to the money aggregates, you will see that the deleveraging process that began in 2008 has ended for the private sector (households plus businesses). I have to think that a resumption of private credit growth is bullish for nominal growth.
8. July 2013 at 13:24
I think the recent surge in yields just margin calls. I’m of the belief that a lot of banks/traders/financial institutions were borrowing short, buying the long end while levering it up and collecting the spread. I think the recent comment by the Fed was the trigger and not the cause. Either way, when the Fed starts to really taper, I believe that interest rates on the long end will actually fall if anything holds. I see deflationary pressures across the world. Japan is basically exporting their deflation while China looks like their unsustainable growth model is about to come crashing down.
I think that Treasuries are gonna go back up in about a year or two. I see deflationary pressures across the board, especially if corporate earnings continue to disappoint. If China experiences a major slowdown(50% of the Chinese economy is investment–NOT SUSTAINABLE) and I think investment levels in China have to drop by around 70-90%, I don’t see how corporate earnings don’t get smashed if China experiences a major slowdown with all of the multinationals that we have. All of those events are extremely deflationary and if it plays out like that, Treasuries are going to shoot up in value. Especially due to the flight to safety aspect.
8. July 2013 at 13:31
Chris, If it’s bullish for nominal growth then it’s not excluded from the model.
9. July 2013 at 16:09
Dr. Sumner:
“Again, where’s that NGDP prediction market?”
Has the market not made “its” decision umpteen times already? If nobody is trading NGDP futures now, it’s almost certainly because the underlying (aggregate spending) is economically worthless. This decision makes sense. After all, nobody can take ownership of aggregate spending. It isn’t an object of economic action.
It is akin to aggregate employment, or the number of fairies on the head of a pin.
Futures markets for oil, sugar, stocks, and so on exist, because market actors value the underlying commodity.
Futures markets, just like every other market, cannot be “designed” by a government. These words would likely cause a short circuit in a monetarist (read central planner) mind, so if it doesn’t make sense now, maybe once you’re old and decrepit, and NGDP has been attempted and failed, just like every other previous central planning attempt has failed, maybe you’ll do what Milton Friedman did in his later life, and admit:
“Any system which gives so much power and so much discretion to a few men, [so] that mistakes “” excusable or not “” can have such far reaching effects, is a bad system. It is a bad system to believers in freedom just because it gives a few men such power without any effective check by the body politic “” this is the key political argument against an independent central bank…To paraphrase Clemenceau, money is much too serious a matter to be left to the central bankers.” – Milton Friedman, “The Money Masters” (1995).
19. March 2017 at 04:22
[…] http://www.themoneyillusion.com/?p=22156 I asked two MMers-Marcus himself and Benjamin Cole what they might think Scott means […]