Do low interest rates stimulate housing?
If you answer this question with a “yes”, then you are reasoning from a price change. I thought of this when reading the abstract to a paper by David W. Berger, Konstantin Milbradt, Fabrice Tourre, Joseph Vavra on monetary policy and mortgage interest:
How much ability does the Fed have to stimulate the economy by cutting interest rates? We argue that the presence of substantial household debt in fixed-rate prepayable mortgages means that this question cannot be answered by looking only at how far current rates are from zero. Using a household model of mortgage prepayment with endogenous mortgage pricing, wealth distributions and consumption matched to detailed loan-level evidence on the relationship between prepayment and rate incentives, we argue that the ability to stimulate the economy by cutting rates depends not just on the level of current interest rates but also on their previous path: 1) Holding current rates constant, monetary policy is less effective if previous rates were low. 2) Monetary policy “reloads” stimulative power slowly after raising rates. 3) The strength of monetary policy via the mortgage prepayment channel has been amplified by the 30-year secular decline in mortgage rates. All three conclusions imply that even if the Fed raises rates substantially before the next recession arrives, it will likely have less ammunition available for stimulus than in recent recessions.
People tend to refinance mortgages when long-term interest rates fall. So what type of monetary policy generally causes long-term interest rates to decline? I’d say the answer is contractionary, whereas the authors of this study seem to assume the answer is expansionary. (I base this assumption on the first sentence of the abstract. I have not read the entire paper, so it’s very possible I misinterpreted their claim.)
This is actually a complex question, and my reading of the evidence is that long-term rates will usually increase when monetary policy is made more expansionary (as in the 1960s and 1970s), but not always. Of course it partly depends on how you define “expansionary”.
Consider the Fed announcements of January 2001 and September 2007. In both cases, the Fed cut rates for the first time in years. In both cases, the policy rate was cut by 0.5%, not the usual 0.25%. In both cases, stocks soared on the unexpectedly expansionary policy news. In both cases, long-term bond yields increased on the news (dramatically in January 2001), even as short term rates declined. If a highly liquid NGDP futures market had existed, then NGDP futures prices would have probably also increased. On the other hand, you can also find lots of examples where short and long-term interest rates move in the same direction. But the two cases I cited are important because they were so easily identifiable–the dramatic market responses at 2:15 pm seemed clearly linked to the Fed announcements. “Identification” of policy shocks is easier in that case.
If I’m right that falling long-term bond yields generally reflect a contractionary monetary policy, then I think it’s a mistake to rely too much on the mortgage refinance channel when the Fed is trying to stimulate the economy.
I believe that monetary policy is always highly effective, even at zero interest rates. We have lots of historical evidence to support that claim. But if it is effective, it’s not because lower interest rates stimulate demand, rather it is because monetary stimulus increases the monetary base and/or reduces base demand, which boosts NGDP. And higher NGDP leads to higher employment in a world with sticky wages. In most cases, long term interest rates will also increase.
HT: Tyler Cowen
Tags:
29. October 2018 at 10:21
Monetary “expansion” & the “crawling economy”
http://ngdp-advisers.com/2018/10/29/the-fomc-deceives-itself/
29. October 2018 at 10:32
Fabrice Tourre? I knew that name sounded familiar:
https://www.wsj.com/articles/goldman-mortgage-trader-convicted-of-fraud-pursuing-new-career-in-academia-1530540001
29. October 2018 at 13:27
The nice thing about TIPs yields is that they account for expected currency debasement, which is part of the ambiguity of nominal rate reactions to changes in Fed policy.
Also, there is the “store up ammo” argument. Google it. Very Serious People, including most bankers apparently, love the “store up ammo” argument, but it seems to me that if you understand that monetary policy can continue to be effective at the ZLB, this argument looks stupid. Thoughts?
29. October 2018 at 13:36
This is an example of where I would apply my “upside down CAPM” model. Real long term risk-free interest rates are an inverse measure of risk aversion. When rates are low, risk aversion is high. So, it appears stimulative, because it does trigger home building. But, that investment is due to a shift from more innovative risky equity investments to investments with more stable cash flows. Similarly, stable, old-school firms with lots of assets have lower cost of capital as a result, so capital is induced to firms like CenturyLink and away from firms like Apple.
Viewing this as a demand-side issue, where lower rates are stimulative, gets everything backwards. Lower rates stimulate stagnation.
29. October 2018 at 13:38
Brian, Yes, the “store up ammo” argument is wrong, indeed it’s even wrong if monetary policy were ineffective at the zero bound. That’s because raising the policy rate reduces the natural rate, and hence reduces “ammo.”
29. October 2018 at 14:24
There are two factors to consider.
If mortgage rates fall, and I considering buying a home, then a lower mortgage rate should make a house more affordable at the same price. If demand is unchanged a lower mortgage rate should be offset by a higher prices. Even so, my thought are the impact from this effect is small.
Based on the excerpted passage, tt looks like they are discussing the impact of mortgage refinancing on aggregate demand. Conventional wisdom says that, if I can refi my 5% mortgage to a 4% mortgage, my monthly payment will drop. This will give me more disposable income to spend on consumer goods, and should stimulate aggregate demand. But, it fails to look at the other side of the transaction. For every mortgage borrower, there is a mortgage lender. The net-net of refi is as wash. That is, unless the borrower is combining their mortgage refinance with equity take-out and levering their balance sheet.
29. October 2018 at 15:13
Doug,
The counterpoint is inflation makes buying more attractive versus renting. Rents are the main component of inflation indexes. So if rents increase 10% every year, then owning a house gives higher future imputed rents. But if you can’t buy with cash, then you also have to pay the higher interest rates.
The relationships between the economy, interest rates and the Fed are very, very complex. Put simply: If the Fed truly meets an inflation or NGDP target, then the Fed does not control interest rates. The economy dictates interest rates the Fed has to set to meet the policy target.
Lower *real* interest rates may stimulate housing, but that’s like saying higher oil prices stimulate oil exploration. If the Fed meets its target, then it has as much control over real interest rates as it has over oil prices.
29. October 2018 at 16:22
“…higher NGDP leads to higher employment in a world with sticky wages.”
Then the 1970s should have had very low unemployment, eh? No stagflation at all. Oh, but of course fans of NGDPLT will plead special circumstances. “Oil!” “Unions!” But go ahead. Re-run the experiment as many times as you want. I’m sure there will always be more ad-hoc scapegoats and epicycles with which to salvage your hypothesis.
29. October 2018 at 16:43
Can you expand on this:
“it is because monetary stimulus increases the monetary base and/or reduces base demand”
(or link to where you do?)
29. October 2018 at 17:58
The entire establishment seems confused by the term “rates”. I wish they’d be more specific.
Do they mean federal reserve rate or Libor? ARMs, which have made a resurgence, are linked to short term rates, and raising or lowering post the reset period has an income effect within 1 year (leaving aside anticipated income). In any case, the wealth effect vs. asset prices (and housing affordability) probably is the more immediate impact – the action is all at the margin.
Do they mean long term rates? Because I’m struggling to see most of the dialogue today making the link, as you note above. There’s no reason that a change in short term rates needs to cause the yield curve to show a parallel shift, and the impact ought to depend on how much slack is in the economy. If the decrease causes expected CPI to go up and the market interprets this as a policy error such that the Fed will need to raise rates more to compensate later, then it could be a wash. If the decrease is during a time of economic slack that is expected to persist (because, let’s say, the decrease was insufficient) then long term rates could go down. Or the decrease could drive up long term rates by boosting AD.
The more meaningful issue – and something that no one seems to be paying attention to – is that unlike prior monetary interventions, the Fed is directly applying pressure to the long end of the curve by letting several billion dollars of bonds roll off every week (forcing the private sector to absorb them even as short term funding costs increase and the Treasury funding needs rise). Ouch. I guess I’m in the camp that the Fed is making a slight policy error but is interpreting the still-steep yield curve as an “all clear”. I suppose we’ll see, but the markets seem to agree with me because they are projecting the Fed will drop rates sooner than the Fed’s own dot plot projects.
In any case, I would struggle with the conclusion that the reduced re-finance incentives are a big deal. In 2016, about a trillion dollars in mortgages got refinanced. A lot of these were ARM resets from 5+ years ealier, so the rate impact was uncertain. If, on average, there was a 1% drop, that’s a net cash flow savings of $10 billion a year. Moreover, with homeownership rates lower the wealth is distributed to folks who already have other assets – and who probably are seeing a much more immediate and larger impact through asset valuations.
However, a 1% increase in rates on a 30 year fixed loan with an 80% loan to value ratio on the property will decrease affordability by 6%-7%. If that manifests into property values (again, the action is at the margin), that affects the asset valuation of $32 Trillion dollars worth of assets. If even 5% of the housing stock trades a year, that’s 6% of 5% of $32T, or about a trillion dollars, and that impacts perceived wealth as well as actual money supply (as debt is paid back to the bank on sale and less debt is taken out to fund the re-purchase). We’re talking 3 orders of magnitude difference.
That’s simple math (but maybe I did it wrong?). Sometimes the folks at the Fed seem pretty far out of touch.
30. October 2018 at 14:00
How exactly does reducing base demand increase NGDP?
30. October 2018 at 21:14
Matthew Opitz, Nobody is saying that higher TREND rates of NGDP growth (as in the 1970s) leads to more employment. Friedman disproved that idea in 1968.
Arilando, For the same reason that more base supply increases NGDP, ceteris paribus. People try to get rid of excess cash balances, and that pushes up nominal spending.
30. October 2018 at 21:45
Off topic, but Brad Setser says China letting the Yuan float and depreciate to a natural low rate would be a big shock to the world (https://www.cfr.org/blog/where-will-yuan-go-next-chinas-big-choice)
That seems a bit off to me. I believe if US threatened to put all Chinese imports on 25% tariffs, one of the best things China could do is to let the currency float and depreciate. It would be stimulative and would benefit others as a stronger Chinese economy imports more than exports. Even capital outflows can help because they bid up the dollar (just what’s needed for a payback in a trade war). The biggest downside risk is the financial system may become unstable, but as long as the government kept NGDP stable, that can be managed. It won’t be easy but certainly doable.
What’s your take Scott?
30. October 2018 at 23:34
Yes, easy money leads to higher rates.
This does raise an interesting question. If the Fed is successful, we should see historically low nominal interest rates for the future.
This means the traditional method of stimulating the economy, lowering nominal rates, will probably be a limited option. We will be boxed in, in the first inning of a recession.
But does quantitative easing work? Seems like it does, but I would not say it is an open and shut case. I genuflect to the QE totem, but sometimes my faith grows weak.
If QE is less effective than we believe (and some people believe QE is nearly inert), then are there other options? Such as money-financed fiscal programs or negative interest rates?
3. November 2018 at 02:12
– Nonsense. It’s the amount of credit that “stimulates” housing.
5. November 2018 at 15:13
If I replace a mortgage with 25 years left with a mortgage with 30 years left, I have net sold duration to the market. From the investor’s perspective, the duration of my 25 year mortgage was low as any further decrease in rates increases my odds of prepayment. In addition, the 30 year mortgage has potential cash-flows 30 years from now.