Scott Sumner to Fed: “There Is No Justification for Listening to My Advice”
I know you guys think I’m just being a mosquito, buzzing around Scott Sumner’s ear and criticizing him for no good reason.
Well you’re wrong. I am the sentinel in the night, guarding the integrity of the USD from Sumner’s attempted debasement, while you have barbecues and watch fireworks.
Now I know very well what Sumner’s apologists are going to say when they hear my latest critique. So, first, let’s consider a hypothetical scenario where a teenager goes to work for a famous chef.
CHEF: In my kitchen, there is no justification for breaking an egg. None.
BILLY: Got it.
CHEF: Now, let’s start with something simple. Here’s a standard cookbook. Go ahead and bake a cake.
BILLY: Okay sure thing. Uh, wait a second.
CHEF: What’s the problem?
BILLY: Well, it says in step #3 that I have to mix two eggs into the batter.
CHEF: What’s your point?
BILLY: Well, I mean, you just told me not to break any eggs.
CHEF: So what? I’m a busy man, kid, get moving on that cake!
BILLY: But, how can I do that without breaking any eggs?
CHEF: *sighs* I meant, don’t break eggs as your direct goal. If you end up breaking eggs as an offshoot of your goal–such as baking a cake–then it’s fine. And in fact, necessary. Now stop being such a smarta** and get to work. But don’t you dare break any eggs. There’s no justification for that, none.
My question: In the ridiculous hypothetical scenario above, would Billy be allowed to complain to his parents/friends that his new boss was either (a) completely incoherent or (b) a jerk playing a trick on him? I think we can all agree that the answer is yes.
If you agree with me on that score, then you should also agree that either Scott Sumner is incoherent, or he is actually Andy Kaufman playing a decades-long prank on everyone. Let me explain.
If you read Scott’s blog regularly, you know that for years he has hammered home the point that it is a stupid mistake to think that low interest rates signify easy money. Scott is quite flummoxed how anyone can think that, since Milton Friedman himself made the point in the late 1990s. Scott repeatedly lectures us to never reason from an interest rate change.
Just to drive home the point, Scott often observes that low interest rates are associated with depressions, while higher interest rates are associated with boom times. When the Fed would put out statements saying it will “accommodate” the economy by keeping near-zero interest rates in place for an extended period, Scott would complain that this was actually telling everyone the economy would stay depressed for another year. (Sorry, I can’t find a link on this last point, but I know he said things like that circa 2011.)
Indeed, if the Fed did the optimal policy–credibly implementing a new policy of level targeting of (market expectations of) constant growth in nominal GDP–then investors would expect strong future growth and the Wicksellian natural interest rate would immediately rise, meaning that the Fed would need to allow the fed funds rate to immediately rise in order to fulfill its new policy regime. But that “hike” in interest rates would hardly be a move toward “tighter money”; it would actually be the cessation of the Fed’s disastrously tight money policy, in place since 2008. (Again, I’m not endorsing all of these statements–I’m just summarizing the worldview that Scott has been painting since 2008.)
OK everyone got all that? Very counterintuitive, but there’s a certain elegance to it. So in that context, how in the world does Scott at EconLog today write this?!
I often see Fed officials claiming that the fall in unemployment means that they need to raise interest rates. Sometimes this is based on “Phillips Curve” thinking—the (false) idea that inflation is caused by a booming economy…
Early in the year there was some indication that wage growth was accelerating. But that no longer seems to be the case, and the latest figures show exactly 2.0% hourly wage growth over the past 12 months. That’s the same rate as we’ve been seeing for the past 6 years, and is too low for the Fed to hit its 2% inflation target. There is no justification for raising interest rates when hourly nominal wage growth is below 2.3% on a 12-month basis. None. [Italics original.]
Now I know, I know, you’re going to excuse Scott here by saying, “Oh Bob, c’mon, he just means, given that the Fed is thinking about things in terms of setting interest rates by adjusting its bond holdings–then it would be a move toward tighter money if it were to hike interest rates. But if the Fed adopts his plan, and promises to buy every asset on planet Earth if need be in order to raise next quarter’s nominal GDP to the targeted level, then interest rates would rise rapidly and that would be awesome. But that’s not really the Fed ‘raising interest rates,’ that’s just the Fed implementing a certain policy that will necessarily result in higher interest rates. That’s how to make sense of his statement.”
OK, but then see again my tale of the Chef and Billy.
Presumably Scott means that if the Natural Rate were above zero at present we’d have faster nominal wage growth.
Don’t ask me how it’s even possible for people to have anti-time preference. I can only conclude that they have schizophrenic expectations about future inflation.
“There is no justification for raising interest rates when hourly nominal wage growth is below 2.3% on a 12-month basis.”
-Ah, Bob, but you forgot what the meaning of the word “is” is. “Is” does not mean “will always be”.
When the Fed adopts an NGDP target, nominal wage growth will not be below 2.3% on a 12-month basis. So Bob is misreading the good and embattled Scott doubly.
Haha, “When”…
“I know you guys think I’m just being a mosquito, buzzing around Scott Sumner’s ear and criticizing him for no good reason.”
-Yup.
“Well you’re wrong. I am the sentinel in the night, guarding the integrity of the USD from Sumner’s attempted debasement, while you have barbecues and watch fireworks.”
-LOL. I actually laughed at that. BTW, I thought the 1970s in the U.S. was optimal monetary policy and was very surprised when Scott strongly disagreed with me on that point. So Scott’s no enemy of the $.
Bob, I think you’ve adequately explained exactly how Sumner’s statement makes sense, but I don’t see what the problem with it is. I don’t think your Chef and Billy example is exactly analogous to the Sumner situation. Here’s an example that’s more analogous (I don’t know much about cooking, so this may not make much sense):
Chef: In my kitchen, there is no justification for boiling water. It’s a useless and unnecessary activity.
Billy: Got it.
Chef: OK, here’s a cookbook. Now go ahead and make some French Fries.
Billy: One problem: it says I should heat up the potatoes. But Potatoes have water in them, so wouldn’t I be boiling water?
Chef: I meant there’s no point in intentionally boiling water. Now if heating up the potato happens to cause water to boil, there’s no problem with that. Just don’t go around boiling water.
If that were the situation, I daresay people wouldn’t object to how Billy was treated.
Here’s an even better example. Suppose a Republican politician was being interviewed by a journalist.
Politician: When unemployment is this high, there is no justification for the government to increase tax revenue. None.
Journalist: OK, so how would you get us out of this recession?
Politician: I would have the government remove all its burdensome regulations, so businesses can more easily create jobs.
Journalist: But if we followed your advice and it got us out of therecession, wouldn’t that lead to the government collecting more revenue?
Politician: Of course it would. There’s nothing wrong with government revenue increasing, it’s just that our goal shouldn’t be increasing government revenue.
Does anyone remember the “somebody else’s business” shield from Hitch-Hiker’s? If you stare at it directly then your subconscious immediately rejects it, not wanting to get involved. So, in order to see the thing hidden there, you must approach obliquely and try to catch a glimpse out of the corner of your eye, without actually trying to do that.
The “Somebody Else’s Problem” field, actually.
Your alternate story doesn’t work. For it to do so, Sumner shouldn’t have said
He should have said
“raising” is conscious, deliberate behaviour, not an unintended consequence of another deliberate action.
Keshav, your hypothetical politician and chef makes no sense. Yes, I would be consistent and say the tax revenue one is obvious, the potato not so much.
However, the reason the potato one is borderline is that people don’t usually think of water in the potatoes. It has nothing to do with the way Scott would have to rescue his interest rate discussion. It’s not as if people don’t even think about interest rates when the Fed has good monetary policy (the way most people wouldn’t even think about water being inside a potato).
OK, ignore my chef example; like I said I don’t know much about cooking. But I don’t see what you find problematic in what the politician is saying. Why doesn’t it make sense to say that you don’t believe in the government collecting more tax revenue during a recession, but if removing regulations gets us out of a recession and ends up increasing tax revenue, that’s OK with you?
Speaking as a chef, if any one of us ever hires you and assigns you a simple first task, and that task is to bake a cake, you are being trolled.
Darien I had to go for brevity. And I thought “scrambled eggs” would be too confusing. If he made the kid chop up vegetables without telling him what they would be for, that would be even more confusing than Scott’s macro writings.
E Harding I can’t tell if you’re trolling me. Right this second, 12-month wage growth is not high enough (in Scott’s view) for the Fed to raise interest rates. You are saying if the Fed adopts LTNGDP then wages will immediately rise. Well, still, that’s the Fed raising rates before the wages rise–or at the same time–and I don’t know that they rise enough immediately to satisfy his requirement.
I expect the central bank to reach its target before rate hikes, but Scott still said “is”, not “will always be”.
No, I’m not trolling.
“But , then interest rates would rise rapidly and that would be awesome”
Are you merely quibbling because Scott said “raise interest rates” rather than “tighten money”? I think in his ideal monetary regime that would be a wrong thing to say – but in today’s world when the fed uses interest rate targeting as its main tool and when they really are thinking about raising rates to tighten it makes perfect sense.
If the fed did what Scott wants and “promise[s] to buy every asset on planet Earth if need be in order to raise next quarter’s nominal GDP to the targeted level’ then both hourly nominal wage growth and interest rates would probably rise. Scott’s statement about it being wrong to raise interest rates when nominal wage growth is below optimal would still be true (just irrelevant to that new situation).
oops, the bit I quoted is irrelevant to my comment (I forgot to delete it before hitting post !)
Transformer wrote:
Are you merely quibbling because Scott said “raise interest rates” rather than “tighten money”?
I would say at least the second biggest point–possibly the biggest point–Scott has been making since he started his blog is that “raise interest rates” and “tighten money” are not the same thing, and indeed usually are not. So I don’t see why that’s quibbling for me to point it out, in a post where Scott is talking about what the Fed should and should not do.
If Krugman all of a sudden came out with a post praising austerity, wouldn’t that be weird? And then if he said, “Oh no, I meant, cut the budget for Guantanamo Bay,” that would be confusing as hell.
That really is quibbling. Obviously, when Scott says the fed shouldn’t tighten interest rates, he means they shouldn’t tighten. He simply says “raise interest rates” instead of “tighten” because the fed is talking about raising interest rates in a manner that would be tightening, and he is talking about why that specific action by the Fed is a bad idea.
He obviously wouldn’t be against the Fed raising interest rates in a way that would be loosening instead.
Analogy:
The Fed is considering turning the steering wheel of a car to the left when the car is already nearing a cliff to the left.
Sumner: “There is no justification for turning the steering wheel to the left when their’s a cliff to the left. None.”
But Sumner has advocated using a self driving car, in which the steering wheel doesn’t turn the wheels, but as a weird side effect, the steering wheel turns to the left when the car turns to the right.
Along comes Bob: “Scott you’re being inconsistent, if we followed your plan the steering wheel would end up turning to the left in this situation.”
But there is no inconsistency, it’s just that the statement “there is no justification…” assumes some conditions that are not explicitly stated in that particular sentence, but nonetheless obvious given a larger context. And objecting to that is, indeed, quibbling.
The reason I see it as quibbling is this.
If Scott’s ideal policy was adopted then the fed would just adjust the size of the base to hit the NGDP target and let interest rates be set by the market. Under this regime an expansion in the monetary base might also lead to an increase in interest rates. Interest rates would be a poor guide to the stance of monetary policy.
Under a regime where the fed does monetary policy by raising and lowering interest rates (Ie NOT Scott’s preferred policy) and people are discussing whether the fed should or should not raise interest rates it seems perfectly valid and not contradictory to say “under the current regime and in the present circumstances I don’t think a rate hike would be a good idea as this would be an inappropriate tightening of monetary policy”, even if you hold the view expressed in my previous paragraph.
To use an analogy. Suppose you’re a fitness trainer who believes that heartbeat is a good way of measuring exercise level during a jogging session, and speed is secondary . You’re training someone who thinks that speed is all that matters. You see their heart beat is too low for a good workout. But given their view you have to say “you need to speed up for a better workout” not “you need to increase your heartbeat for a better workout’ or they will simply reject your advise. But this way of framing your advice in no way contradicts your real views,
How do you envision the monetary base being adjusted without changing interest rates? It’s highly misleading to say that “interest rates would be set by the market” if base money from nowhere is literally being injected into that market by the Government.
But I don’t think you’re reading Scott right if you think he’s only speaking of interest rates for the sake of the “People of the Concrete Steppes.”
Scott is saying that less than 2.3% YoY wage growth is prima facie evidence that the natural rate of interest is, at present, negative. That this is what he’s saying is pretty clear if you read carefully. The problem is he doesn’t really justify this belief. Near as I can tell it comes from assuming people have expectations about future inflation that don’t make sense.
Transformer wrote:
“it seems perfectly valid and not contradictory to say “under the current regime and in the present circumstances I don’t think a rate hike would be a good idea as this would be an inappropriate tightening of monetary policy”
But Bob replied to you just before that:
“Scott has been making since he started his blog is that “raise interest rates” and “tighten money” are not the same thing, and indeed usually are not”.
Don’t mean to be quibbling but that response doesnt seem valid in the context of ssumner
I’m not sure that Bob is correct to say that the point ‘Scott has been making since he started his blog is that “raise interest rates” and “tighten money” are not the same thing, and indeed usually are not’.
Usually “raise interest rates” and “tighten money” WILL be the same thing.
This is is not the same as ‘low interest rates shouldn’t be confused with easy money’ which is Scott’s usual point.
Here would be a more proper story:
CHEF: In my kitchen, there is no justification for breaking an egg into boiling water. None.
BILLY: Got it.
CHEF: Now, let’s start with something simple. Here’s a standard cookbook. Go ahead and bake a cake.
BILLY: Okay sure thing. Uh, wait a second.
CHEF: What’s the problem?
BILLY: Well, it says in step #3 that I have to mix two eggs into the batter.
CHEF: What’s your point?
BILLY: Well, I mean, you just told me not to break any eggs.
CHEF: Well, batter isn’t boiling water. I’m a busy man, kid, get moving on that cake!
BILLY: But, how can I do that without breaking any eggs?
CHEF: *sighs*
“Into boiling water”=”when hourly nominal wage growth is below 2.3% on a 12-month basis”.
As usual, Scott is being 100% consistent.
What in the world are you talking about E Harding? If the Fed implements Scott’s policies next week, do you think everybody will see a raise next week? But interest rates will rise.
So no, Scott is flat-out wrong when he says the Fed can’t raise rates under the current wage conditions. The only out is to say, “I mean, raise rates through contractionary policy, not through my preferred policy.” As I’ve often said, never reason from a price change, unless you’re Scott Sumner. It’s like Willie Mays is allowed to catch fly balls with just his glove, but you tell kids in Little League not to do that.
Bob, why exactly would rates rise immediately if the Fed announced that it would do NGDP targeting? I apologize if I’m asking an obvious question?
Two words: inflation premium.
” If the Fed implements Scott’s policies next week, do you think everybody will see a raise next week? But interest rates will rise.”
Bingo.
Wage increases for every worker lags interest rate increases, as well as price increases, when the Fed inflates.
This is associated with Cantillon effects. Those whose wages increase after price increases, lose in purchasing power. That coercion based transfer of wealth is what we are told is “optimal monetary policy”.
Transformer wrote:
“Usually “raise interest rates” and “tighten money” WILL be the same thing”.
ssumner wrote (quoting MFriedman):
“Low interest rates are generally a sign that money has been tight, as in Japan; high interest rates, that money has been easy”.
Am I missing something or do you and ssumner disagree?
What you’re missing is that Sumner essentially defines the tightness of looseness of money based on NGDP growth. So to the extent that contractionary open market operations (the Fed selling short-term treasury securitities on the open market) would reduce NGDP growth, it would constitute tightening money as per Sumner’s definition. So in that sense a move to raise interest rates in that way would be the same as tightening money.
What you’re missing is that we’re also being told that higher interest rates are “a sign that money has been easy”.
Can’t square a circle.
What is going on here is that Sumner wants to take other people’s definitions of loosening money, for example lowering interest rates, and printing more money, and pick and choose what those phrases mean depending on what is actually going on with NGDP.
If NGDP is actually rising, then he selects the definition “loosening money” for the above phrases at those times, and NGDP is actually falling, then he selects the definition “tightening money” for the above phrases at those times.
In short, he is saying “Raising interest rates IS NOT loosening money if NGDP is falling”, when he should be saying “Raising interest rates SHOULD NOT BE DEFINED AS loosening money if NGDP is falling.”
This is because he believes “It is not useful to say the Fed is being “loose” when millions of people are being throw out of work. It is also a bad idea to cling to old definitions of loose money that could lead the country to another Nazi dictatorship.”
This in turn is because he believes raising employment by way of inflation is a better means than market forces.
Anonymous,
Kershav give a good explanation of why raising interest raises is normally tightening and I think Sumner would probably agree. In addition remember that a CB might raise interest and tighten money in response to a positive demand shock, but if it does not do so sufficiently to keep NGDP on target it might still end up with “loose money” relative to what is needed to hit the target
When Sumner says ““Low interest rates are generally a sign that money has been tight, as in Japan; high interest rates, that money has been easy”, I think he has in mind scenarios where the CB has not lowered interest rates fast enough (or even raised them when is should have lowered) and accidentally engineered an avoidable recession where even at 0% interest rates are too tight to hit the NGDP target.
Ok Keshav I understand that already thanks.
What i meant was bob murphy has pointed out two things relating to ssumner:
1) ssumner told us ‘never to reason from an interest rate change’, yet here he is reasoning from an interest rate change.
2) Using bob murphys kitchen example, ssumner has told us that “There is no justification for raising interest rates when hourly nominal wage growth is below 2.3% on a 12-month basis. None”. (dont break any eggs).
Then when questioned about that says, ‘only if youre targeting NGDP’ (only if youre making a cake).
So ssumner is at one end of the field shouting ‘raise NGDP and dont worry about interest rates; never reason from them’, and at the other end is shouting “There is no justification for raising interest rates when hourly nominal wage growth is below 2.3% on a 12-month basis. None”.
Inconsistent!
Sumner addresses this “inconsistency” here
http://econlog.econlib.org/archives/2014/06/never_reason_fo.html
Where? I don’t see it.
I do see this:
“Actually, explaining 2013 is not more difficult. Monetary policy remained tight, no easing occurred. The fall in interest rates represented the long run effects of the tightening of 2011, just as the fall in rates during the early 1930s reflected the long run effects of a tight money policy adopted in late 1929.”
It is not true the money easing did not take place in 2013. The Fed added to the money supply. NGDP is a faulty indicator. People can choose to spend a little more or a little less than the recent past, which is not a Fed activity, and yet we’re supposed to say the Fed did less or more of something.
Yes, I did indeed slip in the implicit definition that loosening money means inflating the money supply. Yes, I did indeed try to justify that by making an explicit reference to one flaw in defining it in terms of NGDP.
Major Freedom: I think ssumner was referring to monetary policy in the EZ.
Transformer wrote in reference to ssumners ‘low interest rate = tight money / high interest rate = easy money’:
“I think he has in mind scenarios where the CB has not lowered interest rates fast enough (or even raised them when is should have lowered) and accidentally engineered an avoidable recession where even at 0% interest rates are too tight to hit the NGDP target”.
But ssumner writes in the above link you posted:
“nominal interest rates tend to be procylical. They tend to rise during booms and fall during recessions”.
But apparently interest rates don’t matter (never reason from them) instead focus on NGDP and ensure that metric is rising, and if interest rates rise thats great because it means its boom-time. But as bmurphy pointed out how can ssumner then write: “There is no justification for raising interest rates when hourly nominal wage growth is below 2.3% on a 12-month basis. None”.
So what should the policy makers do? Focus on NGDP growth when they are told there is no justification for an interest rates rise. None whatsoever?
But nominal interest rates tend to rise during booms as does NGDP, or so we are told. So ssumner is telling us to raise NGDP but not raise interest rates at the same time (“theres no justification…”). But then at a different time tells us NGDP and rising interest rates go hand-in-hand.
Thats the contradiction bmurphy is alluding to i imagine.