Murphy vs. the Market Monetarists
It is a dirty rotten lie when people say I just go after Krugman. In my latest for Mises.org, I have a long critique of Scott Sumner and Kevin Erdmann’s narrative of the housing boom/bust. An excerpt:
On the face of it, Erdmann is trying to demonstrate that if we use an objective measure, then it seems there was nothing unusual—from a historical perspective—about the growth in the stock of housing in the mid-2000s. After all, even at its recent peak, the particular measure of “Housing Units per capita” was still lower than it had been in the late 1980s.
There are several responses I’ll give to this line of argument. First, who’s to say that the level in the late 1980s was correct? After all, there had been a devastating crash in the real estate market (and related crises in banking) following the “closing of loopholes” in the 1986 Tax Reform Act.
A second problem is that there is an admitted discontinuity in the data set, which is why Erdmann draws the dotted line in his graph. If we just start with the consistent data set beginning in 2000, then the chart is broadly consistent with the claim that there was an unsustainable surge in housing stock in the mid-2000s.
A third problem is that houses nowadays are much bigger than they were in the 1970s. Mark Perry reports that for new homes (of a certain category), living space per person has nearly doubled since 1973. A much more revealing statistic, then, would be something like, “Housing square footage per capita,” which I imagine would have been at all-time highs circa 2007 (though I couldn’t find the data needed to either back up or reject my hunch).
Finally and perhaps most serious, is the problem that Erdmann is playing central planner. We can’t look at aggregate statistics like “housing units per person” and decide whether “too much” or “too little” housing has been built in the country. If we could, then the socialist calculation problem would be a snap to solve.
The only reason why monetarism is even a thing is because of a satanic cult family who controlled almost all central banks around the world.
The reason why there is no “Potatoism” or “TShirtism” is because peaceful exchange of property determines their relative values.
VP Pence’s remarks on administration’s policy towards China:
https://www.youtube.com/watch?v=aeVrMniBjSc
43 minutes, should watch whole thing.
FAKE NEWS: LOOK HERE [RUSSIA], DON’T LOOK HERE [CHINA]
ALL [INSIDE] ROADS TO CHINA ARE BEING CLOSED
RACIST/BIGOTED MONETARISTS ARE BEING EXPOSED
https://theintercept.com/2018/10/09/google-china-censored-search-engine/
LEAKED TRANSCRIPT OF PRIVATE MEETING CONTRADICTS GOOGLE’S OFFICIAL STORY ON CHINA.
https://www.washingtonpost.com/opinions/explain-the-chinese-spy-sen-feinstein/2018/08/09/0560ca60-9bfd-11e8-b60b-1c897f17e185_story.html?noredirect=on&utm_term=.b6a1ff0dc49c
https://www.bloomberg.com/news/articles/2018-10-09/new-evidence-of-hacked-supermicro-hardware-found-in-u-s-telecom
“Western intelligence contacts that the device was made at a Supermicro subcontractor factory in Guangzhou, a port city in southeastern China. Guangzhou is 90 miles upstream from Shenzhen, dubbed the `Silicon Valley of Hardware,’”
What state does [Feinstein] represent?
Austrian view does not explain why the great recession was great as opposed to a mild softening. Housing was crashing for two years and had nearly bottomed before any unemployment at all in the construction industry or the economy as a whole was observed.
https://fred.stlouisfed.org/graph/?g=lufV
If a housing bust was such a killer we would seen problems long before summer of 2008. Housing starts were down by HALF in the summer of 2007 yet the economy absorbed those construction workers no longer working on housing easily.
Sumner is right on this one. Austrian view doesn’t work.
Capt. Parker wrote: “Austrian view does not explain why the great recession was great as opposed to a mild softening”
And yet I worried in October 2007 that we might be in store for worst recession in 25 years.
Housing starts were down by HALF in the summer of 2007 yet the economy absorbed those construction workers no longer working on housing easily.
Did you see the chart in my article? I specifically responded to this charge and yet you’re not even attempting to rebut my point.
My rebuttal is that everything was fine for 50% of the housing cash. Only a tiny bit of softening was seen after 80% of the housing crash. But, then, miraculously, like a critical threshold was crossed, that last 20% of the housing crash FINALLY triggers big construction unemployment and an increase in overall unemployment much greater in magnitude than the decrease in unemployment that we saw during the housing boom. Why there seemed to be a critical tipping point needs an explanation. I don’t see that you addressed that specifically, Dr. Murphy, and that is the problem for the Austrian theory. Sumner’s tight money in the face of a sudden increase in demand for liquidity addresses that threshold event better than the “all about the housing bubble” explanation.
The magnitude of the Great Recession also needs explanation. In your recent article you address it by saying people feel poorer. But why did they feel so much poorer than they felt before the housing boom? Why didn’t they feel only poor enough that GDP followed the path of potential GDP which we now know dipped some, because of malinvestment, post housing bust?
Yes, you did predict a big recession. But when you said:
“From 2001–2004, the Fed kept (real) rates at the lowest they’ve been since the late 1970s. One of the consequences that has already manifested itself is the housing bubble. But a more severe liquidation seems unavoidable.” It seems to me that you were already in agreement that it will take more than the housing bust to cause a severe recession.
Just to be clear, I totally agree that the Fed caused the housing bubble. And I am totally on board with the idea that instead of smoothing the peaks and valleys in the path of GDP, the Fed is causing them. But I also believe Sumner is correct: The Fed made the valley of 2008-9 deeper than it needed to be.
“My rebuttal is that everything was fine for 50% of the housing cash. Only a tiny bit of softening was seen after 80% of the housing crash. But, then, miraculously, like a critical threshold was crossed, that last 20% of the housing crash FINALLY triggers big construction unemployment and an increase in overall unemployment much greater in magnitude than the decrease in unemployment that we saw during the housing boom”
I’m not sure what metric you are using but your description doesn’t square with the Case Shiller price index and the UE rate. Home prices plateau in 2006 and start their decline in 2007, hitting the low in 2012. Eyeballing the Fred Graph there was a 25% decline from the 2007 peak to 2012, 80% of that is a 20% decline, housing prices hit that decline in early 2009- right around April and the civilian unemployment rate was at 9% then prior to a peak of 10% that was hit in October.
Since the low point for the UE rate was 4.4% in 2007 the increase from 4.4 to 9% is ~85% of the total increase in the UE rate came during the first 80% of the home price drop.
I’m using Housing Starts as the metric baconbacon. Please look at the link in my initial post. Housing starts bottomed in mid 2009. Starts were 80% to the bottom in mid 2008 when construction unemployment was STILL FALLING from its off season peak.
Housing starts are not clearly the best metric to use, if you want to use something for correlations between construction winding up and UE then housing completions is clearly a better starting point because that is the point when no more construction workers are being employed by that project.
Housing completions and the UE rate here
https://fred.stlouisfed.org/series/COMPUTSA#0
Starts vs completions are hardly much different going on to mid 2008. Worst case my argument would become: Starts were something like 65% to the bottom in mid 2008 when construction unemployment was STILL FALLING from its off season peak.
Completions 65% to bottom, not starts.
Housing completions did not move in the same way that housing starts did. Housing starts peaked in Jan 2006 and by July 2006 they were at levels lower than every month from May 2003 to May 2006. Housing completions peaked in March 2006 and 6 months later they were still being completed at rates higher than all of 2003 and 2004, 11 out of 12 months of 2005.
Housing starts bottom in Jan of 2009 while housing completions bottom in Jan 2011 and fell by 33% from Jan 2009 to Jan 2011. The decline from Jan 2008 to Jan 2011 was a 60% decline in completions.
The drop from jan 2008 to jan 2011 represents ~47% of the drop in completions from peak to trough.
Additionally there isn’t a linear correlation between starts or completions and employment in construction on the way up in the bubble. From the end of the 2001 recession to the peak in housing starts there is a 40%+ rise, yet total construction employment moved only moved up about 15% and further it only moved up with a lag. roughly halfway into the housing starts surge total construction employment was actually slightly lower than at the end of the 2001 recession. Halfway up the housing construction completion rise and total employment in construction is only up about 10% of its eventual rise.
It’s not easy to break out employment that is specifically related to residential. Much of construction employment is commercial.
“The drop from jan 2008 to jan 2011 represents ~47% of the drop in completions from peak to trough. ”
Correct! Therefore, The drop from the 2006 peak to was 53% but unemployment in construction did nothing from the 2006 peak to January 08 (your preferred date). My initial point was: “everything was fine for 50% of the housing cash” (starts or completions take your pick) So, my mistake. Everything was fine for 53% of the housing crash, not 50%.
“Additionally there isn’t a linear correlation between starts or completions and employment in construction on the way up in the bubble.”.
So, how is it that Dr. Murphy is arguing “Gee, look at construction employment and the housing bubble – see perfect correlation – Austrians Rule! And you, bacon, are arguing “construction employment doesn’t correlate well with the housing bubble so, Austrians Rule!
I claim ownership of the above anonymous post.
@ Capt. J Parker
Murphey’s original post (following the link) says this
“Yet as I explained at the time, there’s a huge flaw in Sumner’s argument. It’s true that housing starts collapsed from January 2006 to April 2008, but that’s not true of construction employment, which fell from 7.6 million to 7.3 million. Even though housing starts fell by 57%, construction employment during that period only fell by 4%.”
There is no “heads we win, tails you lose” situation going on, Murphy and I are pointing out the same fact, which is that new housing starts did not strongly correlate with construction employment.
Here is a recent post I did on construction employment. It is true that the drop in construction employment is a late development. The drop in construction employment is mainly the result of the post-crisis clampdown on lending, which choked off the longstanding migration patterns into growing cities.
https://www.idiosyncraticwhisk.com/2018/09/housing-part-323-construction.html
baconbacon,
Dr. Murphy also said: ” If the recession were about real resource misallocation — as opposed to a nominal spending shortfall — we would expect the unemployment rate to vary according to the fortunes of the housing sector. In other words, if Kling and the Austrians were right, we would have expected the drop in home construction to match up closely with the rising unemployment rate as those laid-off workers tried to find something else to do.” Then, astoundingly, Dr. Murphy goes on to claim because construction employment DID NOT FALL when housing starts fell the Austrians are right anyway. I struggled for a long time with this ” heads I win, tails you lose” argument not believing Dr. Murphy would try to pass off such an obvious contradiction. Finally, I realized that Dr Murphy’s argument probably
goes something like this:
1) Sumner wants us to believe the housing bust was no big deal because resources were easily reallocated and the proof of this is unemployment did not go up starting in 2006 after the housing boom peaked.
2) I, Murphy, claim that resources were not reallocated. The proof that I am right is that construction employment did not fall in 2006 after the housing boom peaked.
The problem with Dr. Murphy’s argument is that he is assuming that reallocation MUST mean reallocation of resources OUT of construction and into something other than construction. Why assume that? Resources could easily have been reallocated to commercial buildings and/or maintenance and renovation. Dr. Murphy’s claim housing starts vs completions makes a difference is wrong. The starts and completions lines are nearly on top of each other.
The Ocam’s Razor conclusion to draw from the hard data that unemployment did not increase from 2006 to 2008 is that the housing bust was no big deal.
@ Kevin Erdmann, I would be very interested in your opinion as to the likelihood that post 2006, construction workers were reallocated away from new home construction and into other construction jobs (as opposed to remaining in new home construction despite the decline in housing starts or as opposed to leaving construction.)
Capt.
Yes. That is the point of the Dean Baker post that I discuss in the post I linked to. To some extent it seems clear that that was happening.
Baker makes the conventional presumption that Bob does, that this was just one bubble replacing another, but as I point out in the link, both rental income and capital gains for commercial construction in multi-unit residential have been strong, even measured from 2005.
@ Capt J Parker
I don’t believe that Murphy every mentioned housing completions, that was me, and I was pointing out that the concept of using housing starts is a bad idea even if you are just talking about housing and construction employment on their own. The point of ABCT is that you have to look at the structure of capital investment to figure out what is going on, using starts and not completions + employment is missing an obvious point about capital structure.
Now I am talking for myself, and not Murphy, but I read this piece as an indictment of SS not getting capital structure. What SS is doing is erecting the weakest possible argument (without even full quotes or attribution but arguing against unnamed commentors) without it being a literal strawman. He defines the housing bubble as only being about single family residential starts and the bust as the drop off in starts, not Murphy. If you follow the link back to the 2011 piece Murphy makes this point again, that housing starts doesn’t correlate strongly with construction employment (I think he should use total construction workers employed with total housing starts rather than a UE rate).
BTW SS is criticizing Kling here is explicitly NOT focusing on ABCT as he says
“Austrians want to focus on changes in the capital intensity (or “roundaboutness”) or production as the typical (only?) shock that can affect the pattern of trade. I will grant that in practice this can be important, but I do not wish to make it the central issue.”
I could have saved the time and written the following.
As it stands the monetarist critique in this line of the ABCT position can be almost entirely reconciled by changing the term “housing boom” to “construction boom”. At best it is an (insightful) observation that Austrians were focusing to narrowly on one sector and missing a broader, but still consistent with their theory, picture. At worst its a deliberate misrepresentation of ABCT to impugn a concept by framing the debate with semantics rather than technical accuracy or understanding. I have a personal opinion on where SS falls in this continuum, but other than that the critique has no teeth.
baconbacon,
Just to be clear, the position I am defending is that the Great Recession was Great because of a financial crisis that the Fed had power to avert and not because of a large, real resource misallocation from the housing bubble. Single family housing starts might not be the best measure of construction activity but it is a measure. I claim that the lack of correlation between housing starts and construction employment as proof that reallocation of resources that were once used to build house was no problem for the economy in 2006-2007. If I am wrong because housing starts is such a poor measure of housing bubble construction activity then what is the correct measure? It has to be a measure other than employment. If the argument is that employment is the only valid measure of efficient sustainable real resource allocation then the theory that the Great Recession was because of unsustainable real resource allocation can never be disproved. The Great Recession was caused by resource misallocation becomes a tautology.
@ Kevin Erdmann
from the piece you linked
“But, it was only after the financial crisis that construction employment saw its steepest decline. The recession caused the contraction in construction employment.”
This is at best a misleading conclusion and not supported. Total construction employment in July 2007 was 7.66 million people, in December 2007 it was 7.49 million. It is technically incorrect to state that the recession caused construction UE as it was already declining and the logical causal arrow goes the other way, that declining construction employment was a cause of the recession. Then there is a further decline to 7.11 million employees by August 2008. So construction was shedding a little under 50,000 jobs a month from December to August, and that number had accelerated from 25-30,000 a month in the 6 months preceding the recession. The slope is steeper still if you go from March to August/September where jobs are being lost at 70-80,000 per month.
All of this occurred under the umbrella of stable inflation, stable inflation expectations and Fed loosening.
About half a million construction jobs were lost from July 2007 to the point where there is any credible claim (by monetarist definitions) of tightening. This next bit requires date selection, and I am going to use the end of the recession as a neutralish starting point (its not intentionally cherry picked).
From the end of the recession to the bottom of construction employment saw a further reduction of around six hundred thousand jobs. This period is characterized by forward looking inflation measures being almost entirely above 2% (short dip in 2010 below 2%, does not correlate with an acceleration in construction losses at all) and a good amount of time above 2.5%.
This puts fully half of all construction job losses under a period of 2%+ inflation expectations with job losses preceding the recession. The only way that you can claim that the majority of job losses were caused by Fed tightening is to argue that job losses in construction would level off on their own despite the fact that they were accelerating at that point anyway.
That is of course an easy statement to make, but it is a very difficult one to come up with evidence for.
@ CJP
” If I am wrong because housing starts is such a poor measure of housing bubble construction activity then what is the correct measure? It has to be a measure other than employment.”
As I said housing completions is one step in the correct direction. If you take total construction employment and housing completions they peak at near identical times. Total construction employment had been rising at a good pace, ~15% from March 2003 to March 2006. Stalling out those levels of growth is a notable event.
This is also the peak of housing prices, and the housing bubble is about the increase in construction along with the increase in prices. Housing prices and new construction rose while the vacancy rate rose, this is a very bizarre situation.
My bad again. The above is mine.
Hjalmar Schacht, the Reichsbank and the Rentenmark.
https://en.wikipedia.org/wiki/German_Rentenmark
Modern USA banking has a big component of mortgage backed collateral which is part of what supports the US dollar (or FRN if you prefer). The institutions of the FNMA and GNMA encourage this outcome, but also as we know gold is a bit barbarous. Now when you choose something as backing for your currency, and a large chunk of your banking sector, it’s kind of important you choose something stable.
Mortgages are handy, because most of the time, most people always pay their mortgages thus creating a reliable income stream, and there’s a solid tangible asset to repossess and sell in the fallback situation. Problem is when the housing sector starts to swing wildly, suddenly quite a lot of people stop paying their mortgages and the asset prices fall across the board. Oh dear, now the bum has fallen out of our entire payment system. What to do?
I’m not arguing against what Murphy is saying about the problems inside the housing sector totally agree there was too much investment in that sector and also a deliberate government policy of forcing banks to lend to low quality borrowers, as well as using GNMA to write insurance for mortgages that should not have been insured, and using FNMA to provide a subsidized buyer that artificially pulled money into one particular sector. And yeah, the Fed was at least partly to blame for low interest rates, but personally I blame other factors more than the Fed. What I’m saying here is that problems in the housing industry were only the start of what went wrong.
Unemployment in construction was at the leading edge, but if you look at something like Bear Stearns and Lehman Brothers, they had essentially made huge bets on the housing market and also they had ties into all sorts of other areas of the financial industry. So that’s the linkage that allowed the housing shockwave to hit everything else (with a slight delay).
The bit that Sumner can’t get his head around is that all currency (and the entire financial industry) depends on that old “confidence fairy” that Krugman used to talk about. Without the fairy helping out, you can’t redeploy resources because no one trusts that the currency is properly backed. Well, eventually there’s always ways to sort it out, but it does take a bit of time. So the Fed had no choice but to impose some level of discipline on the banking industry but also to at least partially bail out the mortgage sector, if they wanted to maintain their position of a trusted currency issuer. Not everyone went away happy, some moved to gold, others moved to Bitcoin, but you never make everyone happy in these situations. Point is that the US dollar did not crash, people still trust it, and the Fed largely nuanced their way out of trouble (so far).
Sure, but you were predicting an inflationary spiral and that didn’t take into account the large debt-deflation taking place in the mortgage industry, which is what the Fed was trying to compensate for with more liquidity. If the Fed had done nothing we would have got really deep deflation as many mortgages unwound, and all the financial leverage on top of those mortgages also unwound. Maybe that would have been a good healthy lesson in not leveraging yourself too hard with ultra loose lending standards, but anyway the Fed decided to water down that lesson.
You don’t really need to look to ABCT to explain the housing boom/bust. A simplified but common sense view might be:
– Low interest rates and lax lending created a high demand for housing leading to expanded construction and higher house prices
– By 2006/7 many of the people who had taken our loans they couldn’t afford started to default and the housing boom stared to bust
– The financial institutions that made risky loans to the people who were defaulting , or bought assets based on these defaulting loans . themselves went bankrupt or would have done if they hadn’t been bailed out.
– The major disruption to the financial markets caused by the bankruptcies and bailouts led to the biggest recession in 70 years.
If this was ABCT at work I would have expected other sectors than housing to show signs of mal-investment. Even within the construction sector I would have expected that more “roundabout” types of building like sky scrapers would be more prone to mal-investment than simple domestic housing where most of the alleged boom appeared to be centered. Is there evidence of these things happening ?
Here’s Rothbard talking about normal free market speculation, vs “mass error”:
So the critical difference is the “mass error” when large numbers of people make the same mistake at the same time. This requires some kind of central coordination, and Rothbard was fingering the Fed in the context of the 1920’s money supply expansion, followed by 1930’s unwinding.
However, if you go back to first principles, there’s no particular reason why the Fed would be the ONLY central coordinating body, nor even necessarily the most significant. Looking at FNMA, they publish “Loan Limits for Conventional Mortgages” as a national standard as to what they consider to be a reasonable house price (by location). This isn’t price fixing but it is strongly influential in the industry.
They also supply a range of online tools, that although not compulsory, do tend to get used partly because it’s easy, and partly because if you want to be able to resell your mortgage into the secondary market you need to make sure you are aligned with the FNMA standards.
https://www.fanniemae.com/singlefamily/originating-underwriting
Thus we have a strong system of central control in the mortgage market, largely by one government regulated entity being the most powerful buyer by far in the secondary mortgage market. This significantly increases the likelihood of “mass error” because instead of many and varies speculative activities going in every possible direction, we have one big leader and many followers all following an identical recipe.
Don’t blame the Fed all the time, there’s blame to go around without needing to step outside the Austrian Economic framework.
Capt. Parker wrote: “Austrian view does not explain why the great recession was great as opposed to a mild softening….If a housing bust was such a killer we would seen problems long before summer of 2008. Housing starts were down by HALF in the summer of 2007 yet the economy absorbed those construction workers no longer working on housing easily”
Transformer wrote: “If this was ABCT at work I would have expected other sectors than housing to show signs of mal-investment.”
OK so no matter what, the Austrians were wrong. There should have been problems only in housing (according to Sumner/Parker) but it was a general downturn, so Austrian story doesn’t work.
But according to Transformer, the Austrian story is wrong because we only saw problems in housing.
I agree I can’t beat your guys’ arguments on this one. One of you will prove me wrong.
You seem to have missed the fact that I was framing my comment as a question. Hence my final sentence ‘Is there evidence of these things happening ? ‘
And in any case the standard ABCT story would have both malinverment in specific areas and (eventually) a generalized downturn so I really don’t know what to say about your response.
Transformer wrote: “You seem to have missed the fact that I was framing my comment as a question.”
Am I taking crazy pills? I’m just asking.
I don’t know about crazy pills, but there is evidence you are taking grumpy pills!
I was genuinely asking a question about supporting evidence for ABCT, but nevermind.
Tone of voice is hard to assess in print, Transformer, so I may have erroneously fired in self-defense.
Its alright. I’m clearly not blameless on the tone front.
Its probably as good a time as any to say that this is by far my favorite blog and I have enjoyed and learned loads from the discussions here.
Transformer wrote: “Its probably as good a time as any to say that this is by far my favorite blog…”
The important thing is that you like it more than Gene’s.
I’ve definitely learnt lots about tone of voice from Gene’s replies to his commentators.
If the great recession were all about malinvestment not only in the housing sector but also in the rest of the economy to an extent rivaling the size of the housing sector then massive money creation by the Fed along with ARRA should have caused the inflation you were expecting, Dr. Murphy, since the full size of the downturn would have been a supply problem. There was no inflation.
@Capt. J Parker
I’m putting this comment here because the nesting irritates me and because some of what I will write is tangential to the conversation. These are my views and might not be consistent with Bob Murphy’s interpretation of events and ABCT.
When I look at SS’s claims I think what he is saying is that the general downturn due to the lending bubble was not large enough to produce the great recession and that the timing of the fed’s Lehman (in) actions represent some type of tipping point that convinced the market that things were going from tough but mostly OK to everything is going to be a disaster. My response here to this (overly simplified version) is that
1. Delinquency rates on commercial real estate loans started moving up in late 2006 and started steepening in late 2007, and there is no obvious increasing of the rate of delinquency growth in the 2nd or 3rd quarter of 2008.
https://fred.stlouisfed.org/series/DRCRELEXFACBS
2. Single home mortgages show a similar pattern.
https://fred.stlouisfed.org/series/DRSFRMACBS
Single home mortgage delinquency rates had doubled or more from Q2 2006 to the very beginning of the recession at the end of 2007 and more than tripled to Q3 2008. Commercial real estate delinquencies tripled from Q2 2006 to the beginning of the recession and more than quadrupled to Q3 2008. Both of these metrics appear to either lead the UE rate and raw employment numbers by a quarter or happen coincidentally with them, and the early (pre Q32008) rises in delinquencies far exceed the increases in UE.
This indicates that the delinquencies were not caused by rising UE. Measures of AD, inflation and inflation expectations are all normal prior to the recession and not flashing red in any way prior to Q32008. Further the Fed cut interest rates from July 2006 to July 2007 by a significant amount and the Federal Government push through a 100+ billion dollar stimulus bill in Feb 2008.
“Further the Fed cut interest rates from July 2006 to July 2007 by a significant amount”
No, that’s wrong. The effective fed funds rate was constant during that time period.
https://fred.stlouisfed.org/graph/?g=lwLx
“This indicates that the delinquencies were not caused by rising UE”
Well yeah, they were caused by the interest rate increase by the Fed that began in 2004. Once ARMs reset many people couldn’t afford their mortgage anymore.
“Measures of AD, inflation and inflation expectations are all normal prior to the recession and not flashing red in any way prior to Q32008.”
One thing that was flashing red prior to the recession was the TED spread. This is what prompted the FED to start cutting rates in 2007. By 2008 inflation expectations had jumped up so the FED stood pat at a 2% Fed Funds rate comforted, no doubt, that the TED spread had abated some by mid 2008. But as you point out, delinquencies continued to rise. Sumner’s claim is that NGDP was dropping and there was still rising money demand in order to cover bad loans in the banking system but the Fed stood pat in 2008 at 2%. It it this refusal by the Fed to cut rates further that was the tipping point, not Lehman, which was later. The 2008 stimulus was pouring water on smoldering ashes and expecting that will rebuild the house that the Fed torched.
How are you linking Fred graphs with multiple lines?
“No, that’s wrong. The effective fed funds rate was constant during that time period.”
This was a typo, rates rose form July 2007 to July 2008, I don’t believe that typo changes the thrust of the comment.
“Well yeah, they were caused by the interest rate increase by the Fed that began in 2004. Once ARMs reset many people couldn’t afford their mortgage anymore.”
This is probably of limited importance. The most common ARM was a 5+1 year ARM, reset wouldn’t happen until 2009 for a 2004 loan and full reset wouldn’t happen until 2010. The Fed started raising rates in 2004 and delinquencies took a long time to start increasing, into 2006 after rates had risen almost 5 full percentage points. Further more the 30 year and 15 year moved only a modest amount with the increase in the funds rates. Just before the first funds rate rise the 15 year was at 5.7% and the 30 year was at 6.3%, at the end of the hikes the 15 year was 6.3% and the 30 year was 6.7% and right before the next round of decreases they were still at 6.3% and 6.7% and those were the peaks. If you had a 2 or 3 year arm starting in 2004 you would have seen roughly a 0.4% interest rate increase when you refinanced into a 30 year compared to what a 30 year would have cost when they took the loan out. Fred data on the 5 year arms only goes back to Jan 2005, but then it was at 5%.
If you took out a 5 year arm in July 2001 and were “stuck” refinancing 5 years later in July of 2006 you actually were faced with lower 30 and 15 year mortgage rates than in 2001 (this rate might have been higher than the ARM rate you got) and this goes for (I believe, eyeballing the graph) every 5 and 7 year arm taken out starting in late 2003. (I was actually one of the “unlucky” ones, bought a house in the summer of 2003 with a 5 year ARM and rates were almost exactly the same come refinance time in 2008, on the other hand we had a $150,000 loan at half a point lower for those 5 years and around $4,000 in interest payments).
“One thing that was flashing red prior to the recession was the TED spread. This is what prompted the FED to start cutting rates in 2007.”
Ok, the TED spread starts widening in 2007, the Fed starts cutting. Did the TED spread narrow? The Fed cut the funds rate by 2.5 points from July 2007 to July 2008, the TED spread in July 2007 was 0.5 and in August was 1.7, in 2008 it was 1.1 and 1.1. The best case scenario is that 2.5 points worth of funds rate cutting shrank the TED spread by 0.6 points (worst case is it increased it by 0.6 points), to push the TED spread down to 0.5 again the Fed would have had to been at ZIRP in July 2008, and inflation expectations were rising during this period.
“Sumner’s claim is that NGDP was dropping and there was still rising money demand in order to cover bad loans in the banking system but the Fed stood pat in 2008 at 2%. It it this refusal by the Fed to cut rates further that was the tipping point, not Lehman, which was later. The 2008 stimulus was pouring water on smoldering ashes and expecting that will rebuild the house that the Fed torched.”
What Sumner conveniently forgets to mention is that his preferred framework for monetary policy is a combination of using forward looking metrics PLUS catching up for previous mistakes. Sumner has also stated that in the absence of a nGDP futures market he prefers the Fed target forward looking inflation.
If we take this framework and the facts at hand- that is rising inflation expectations which hit (by the metric you linked) over 5% in June and July 2008 then the Scott Sumner recommendation is to TIGHTEN. Even if you look at other inflation expectations they don’t drop until late September, that the Fed stopped lowering rates in July does not seem to have effected the market. The TED spread in August at 1.1% is lower than most of the preceding 12 months and also doesn’t move up until mid September.
“to push the TED spread down to 0.5 again the Fed would have had to been at ZIRP in July 2008, and inflation expectations were rising during this period”
Yeah, the Feds focus on inflation worked out really well didn’t it? It’s really a good thing they stayed away from ZIRP for for those two months because, because, um, hmmm – I’ll get back to you on that one…
So exactly what framework would the Fed have to use for them to have been able to predict that inflation wasn’t going to be a problem? Retroactively deciding what policy should have been doesn’t work with monetarism. Your snark is at least one level above SS who refuses even to acknowledge that by his own preferred metrics the Fed should have been tightening EARLIER than it did, his framework only functions if he is allowed to choose the dates to start employing it, and to ignore some metrics. The Fed doesn’t have those luxuries.
baconbacon,
I’m not going to answer your question. The FED has a huge staff and is supposed to look at everything. Ted Spread, mortgage delinquencies, leading GDP indicators, Bear Sterns bailout in early 2008, everything. I claim they got it wrong.
BUT, I though we were debating if this was a real resource issue vs a monetary policy issue. The fact that the Fed had a difficult job in determining which indicators to pay the most attention to does not prove the great recession was great because of a real resource issue.
@ CJP
You are starting from the assumption that the Fed could possibly have gotten it right, The straight forward conclusion from rising inflation expectations with rising unemployment is that the Fed could not get it right in a meaningful way. Either they choose to tamp down inflation and take the hard swerve down or they ignore inflation and take the repercussions of stagflation or worse.
“The straight forward conclusion from rising inflation expectations with rising unemployment is that the Fed could not get it right in a meaningful way”
No, that’s totally wrong. Too high inflation was never the problem.
https://fred.stlouisfed.org/graph/?g=lysq
All you show here is that to high inflation isn’t a problem as long as the Fed made sure it wasn’t a problem. This is cake and eating, you don’t get to say “the fed was to worried about inflation and acted because of it” and then assume that inflation would be totally fine without that action. Taking just your graph the two other times that inflation expectations by the UoM methodology spiked in that graph it was followed by core inflation increasing by about 1 percentage point. The 2008 spike is both higher and persists for longer, and actual core inflation at the time was higher than either of the other two episodes. Eyeballing just these numbers would lead you to guess that Fed loosening at the time would push actual inflation to the 4.5-5.5% range
Your position, and SS’s, relies on switching between actual and expected rates when convenient. Actual core inflation doesn’t drop below 1.7% until after the recession is over and doesn’t hit its bottom until more than a year after the recession.
You don’t get to use actual inflation in 2008 and say “see, no inflation worries” because then I get to say “but inflation was fine until the end of the recession so tight money is clearly not the issue in 2008”.
“then I get to say “but inflation was fine until the end of the recession so tight money is clearly not the issue in 2008”
If 2% inflation with 9% unemployment is your idea of “fine” then I Beseech You, in the Bowels of von Mises, Think it Possible You May Be Mistaken.
@ CJP
You are dodging the point. Your statement that “to high inflation was never the problem” relies on taking inflation measures alone and ignoring expected inflation. The position that the Fed was to tight relies on taking expected inflation and ignoring actual inflation. You do not get to build a coherent model where you swap back and forth between measures post hoc, it won’t work.
Using the same framework that SS himself uses to claim that the Fed was to tight in August/September/October 2008 gives you the result that the Fed was to loose in Jan through July 2008.
I am not stating that 2% inflation with 9% UE is fine, I am stating that the Fed has to act using market data and the market data that you are choosing to support your position in time period X implies the exact opposite response in time period X-1.
“The straight forward conclusion from rising inflation expectations with rising unemployment is that the Fed could not get it right in a meaningful way.”
I interpret this statement as saying no matter what the Fed did it would have failed at keeping unemployment from getting as high as it did. The fact that inflation remained low is proof that the Fed never really tried.
If what you meant instead was that there was no reason at all for the Fed to adopt a looser monetary policy in early 2008 knowing what they knew at the time, I reject that assertion. The Fed just finished bailing out Bear Sterns, TED spread was high, The Fed had inverted the yield curve and was trying to undo that damage, consumer sentiment had crashed, Leading indicator index was on the verge of going negative: https://fred.stlouisfed.org/graph/?g=lyZO
And even if I am totally wrong and only a madman would have adopted a looser monetary policy in 2008 that still does not prove the Great Recession was a real resource allocation problem. The fact that all the stimulus and money printing after 2008 STILL did not cause inflation proves just the opposite.
” It it this refusal by the Fed to cut rates further that was the tipping point, not Lehman, which was later”
I am not sure what you mean by this. Lehman dropped by 90% on September 15th, the largest market down legs came after this. The inflation expectations measure you linked was still at 4% in October, The TED spread doesn’t start rising again until September 11th/12th which is when Lehman rumors were flying.
” It it this refusal by the Fed to cut rates further that was the tipping point, not Lehman, which was later”
I am not sure what you mean by this. Lehman dropped by 90% on September 15th, the largest market down legs came after this. The inflation expectations measure you linked was still at 4% in October, The TED spread doesn’t start rising again until September 11th/12th which is when Lehman rumors were flying.
Sorry, left out this very busy Fred chart with Fed Funds, TED spread and all that.
https://fred.stlouisfed.org/graph/?g=lwOz
From reading https://mises.org/library/my-reply-krugman-austrian-business-cycle-theory I’m left with the impression that the housing boom was just a housing boom, not a general long-term projects boom, as you would expect from the ABCT.
Why was housiing and just housing the epicenter of the boom and bust? Or wasn’t it?
If it was just housing, couldn’t we explain it (or at least conceive of a different hypothetical scenario) without interest rates even changing? Imagine that the government prints money and uses it to pay companies to build houses — or creates a special lending program just for houses, but don’t messes up with the general interest rate -, wouldn’t that have basically the same effect?
If so, perhaps we should start considering a new ABCT version that just talks about new money being created and going to specific sectors, instead of the whole interest/intertemporal adjustments/hayekian triangles talk. Why is this wrong?
Any time new money is created it always goes somewhere, and highly unlikely to exactly spread equally every direction.
The money creation process tends to be rigged towards government borrowing and that’s the payment the banks make for their protected cartel status. For example, at the moment Australian government 10Y bonds have an interest rate at less than 3% PA while there is no person or business anywhere in Australia that can borrow money at a fixed rate of less than 3% over 10 years. Typical mortgage (even if you offer massive collateral such that risk is zero) would be around 5% and only fixed rate for a few years.
It’s always rigged in the government’s favour because they can push the banking system to buy up those government bonds. Governments can also create special regulations to “nudge” money in certain directions (e.g. in Australia they put a compulsory fraction of your income into retirement funds which have very strict compliance rules as to where the money can go).
It’s called a Cantillon Effect. As I was trying to say above, you can get an “Attack of the Gnomes” type situation (or “mass error” as Rothbard calls it) triggered by any kind of systematic price distortion, be that spacial or temporal (or a bit of both). It’s probably more accurate to say that the gnomes were there all along but keeping themselves invisible while they mess things up, suddenly one day everyone can see the gnomes and reacts accordingly. FNMA and GNMA are essentially gnome attack HQ.
https://mises.org/library/how-central-banking-increased-inequality
You will find similar concepts explained in many places.
When a model is a century or more old and hasn’t been taken seriously in roughly a century, it’s probably wrong. If Austrians haven’t been able to convince the majority of economists by now, they never will.
Otherwise, the market for ideas is extraordinarily inefficient indeed, which would undermine the Austrian market fundamentalism anyway.
“When a model is a century or more old and hasn’t been taken seriously in roughly a century, it’s probably wrong.”
I highly recommend thinking for oneself using logic and evidence, rather than following the crowd or popularity contests, as the better means to separate truth from fiction.
NPC or human?