Recent Posts

Monday, November 22, 2021

Interest Rate Policy Versus Alternatives

One of the ongoing arguments about Modern Monetary Theory (MMT) that I run across is the general disdain for monetary policy among MMT proponents. (At one extreme, Warren Mosler argues that interest rate policy works in a way that is backwards versus the consensus.)

Interest Rate Policy Ineffectiveness

The MMT position is straightforward, yet critics seem incapable of framing it correctly. Although Mosler is not alone in his views, many academics have a more nuanced position: interest rate policy has mixed effects, and is much weaker than the mainstream assumes is true. (This is in line with many post-Keynesians, although post-Keynesians are all over the map in terms of specifics. I believe it is safe to say that some post-Keynesians have views about monetary policy that are extremely hard to distinguish from conventional ones.) Whether or not MMT proponents’ policy prescriptions have merit, the argument that interest rate policy is largely ineffectual is a stand alone area of debate.

To quickly recap Mosler’s argument, a good portion of it relies on the basic accounting fact that raising interest rates increases the interest payments on government debt. Funnily enough, if you spend any time looking at the incoherent mess that is neoclassical analysis of fiscal policy, this actually should not be controversial. Neoclassicals love warning about “debt spirals” and “fiscal dominance” — which is what Mosler argues, except that they present it in with the maximum amount of obfuscation possible. The obvious explanation for the obfuscation is that this observation blows a hole in the conventional logic about interest rate policy. The neoclassical view is that increasing interest rates lowers inflation — except when it doesn’t.

However, the interest rate expense channel is not the only thing driving the economy. This is where the “mixed effects” comes in. In my case — which may or may not reflect other MMT proponents — I am sensitive to the housing market in the “anglo countries,” and the housing market is interest rate sensitive.

The problem for analysis is that the housing market is indeed a market. Neoclassicals love setting up models with “step/impulse responses” (as per control systems), and so they can make scientific-y statements like “a 25 basis point rate hike lowers the inflation rate by 7.3547 basis points in the following quarter.” The problem with the housing market is that nobody would treat such a statement seriously. House prices act like other risk assets — they typically march up steadily, until they drop like a rock. At best, one can hope for the Goldilocks scenario of a “soft landing.” (When you are describing your policy outlook using a character from a fairy tale, it is a safe bet that you are not working with a settled science.)

Canadian Example

Chart: Canadian Construction Employment %

The situation in Canada offers a good example of the box that New Keynesian central bankers constructed around themselves. Construction employment (figure above) has marched to high levels relative to past history — despite somewhat sluggish population growth. Admittedly, there was considerable under-investment in infrastructure in the 1980s and 1990s in Canada, and there has been a catch-up effect in non-residential construction. Nevertheless, residential construction is perky, and a major driver of economic growth.

Chart: Canadian Household Debt Service Burden

I was never too happy with publicly available Canadian house price data, but by all reports, house prices have been taking off like a rocket. This is not entirely an accident, as interest rates have being confounding the bond bears and steadily marched lower for decades. The chart above shows total debt service expenses for Canadian, mortgage and non-mortgage. (I am working from memory, but I believe that lines of credit are in the non-mortgage service component, but the only reason banks are so generous with them is that they know that borrowers have housing assets.)

The debt service burden has been stable. This is the result of interest rates dropping, as well as the reality that only a small percentage of the housing stock turns over in a given year. Anyone who bought housing years ago is in much better shape than new buyers.

The concern is that Canadians (unlike the Americans) cannot lock interest rates for 30 years. The de facto maximum rate lock period is 5 years, after which the interest rate terms of the mortgage needs to be rolled over. A secular increase in interest rates would make the debt service chart ugly. This means that the “pick a policy rate level out of thin air” — very popular among the commentators that I ignore — guesses about interest rates (e.g., 6%) tend to end up much higher than realised outcomes.

What are the Alternatives?

MMT critics love picking out MMT statements about taxes from primers, and assume that the only policy lever available is tax policy. This means that if Canada wanted to cool the economy, a tax hike is allegedly the only option.

This is not even remotely correct. The Canadian housing bubble was launched by monkeying around with the Canada Mortgage and Housing Corporation (CMHC) limits on mortgage insurance. (Under Canadian law, any mortgage with a loan-to-value greater than 80% must get insured.) If policymakers wanted to cool the housing market, all they need to do is move the mortgage insurance limits back to more sensible levels. (They did make steps in that direction, which helped slow housing in the 2010s).

Such a policy has obvious risks, but so would rate hikes. Meanwhile, adjusting lending standards would be a fine-tuned policy aimed exactly at a problem area within the economy.

To what extent it matters, my view is that inflationary pressures are largely transitory, and so I am unconvinced about the need to tighten policy. The housing market is overheating, and it is probably time for responsible adults to dunk that market in cold water. As noted, we do not need interest rate policy to do that. A broad-based tax hike would only be needed to deal with widespread inflationary pressures, which I am unconvinced about.

Email subscription: Go to https://bondeconomics.substack.com/ 

(c) Brian Romanchuk 2021

15 comments:

  1. I don't know if it is the same in Canada as the UK, but today I got a note talking about a 40 year mortgage.

    40 years.

    So when they do get around to putting up interest rates, it won't have any effect - because the lenders will just push the term out. It'll all be about keeping that monthly payment at the maximum 'affordable' level regardless of the interest rate. That's what people look at. They haven't a clue how interest rates work.

    If we want to deal with house price bubbles, then reducing the retirement age could help, and failing that proscribe the term length directly.

    Warren's position is that there is no justification for public usury. Looking at the history of how this came about, prior to 1620 usury was considered a mortal sin and banned by the church. English kings, for example, demanded and obtained interest free loans by issuing anticipation instruments. Those 'forced loans' (the equivalent of today's bank reserves in the 'forced loan' department) changed into 'commercial loans' about the middle of the 17th century with the rise of the banking class and the natural value of money belief of John Locke and co.

    MMT shows that the tax power can be used to re-establish the 'forced loan' system. And since such a system has worked in the past for several centuries, there's no reason it won't work today.

    Public usury serves no purpose other than to enrich bankers and financiers. It's time we stopped doing it and took a different approach. More tightly regulating the quality and type of loans banks can make for example.

    Increasingly I find the economist's obsession with interest rates peculiar. If the only tool is an interest rate, everything ends up looking like a loan.

    ReplyDelete
  2. I think most of us know there are alternatives to interest rate policy. The better question is can politicians who must seek re-election be trusted to use them. Seems to me the answer is "no."

    Central banks are independent, can move quickly, and are less subject to the whims of politics which is whey we outsource monetary policy to them. Perhaps we should give them more tools.

    As for Mosler, he may have some soul-searching to do after recommending Turkey resort to lower rates to firm up the currency and cool inflation. His recent silence on the topic speaks volumes.

    ReplyDelete
    Replies
    1. 1) The changes to CMHC policies were by unelected bureaucrats.
      2) If interest rate policy is ineffective, it doesn’t matter how they are set.

      Delete
    2. on 2) I think the burden of proof is pretty high to make the claim that interest rate policy is ineffective. No doubt in my mind that if the Fed, for example, did a surprise hike tomorrow that inflation expectations and commodity prices would fall hard.

      The better argument is that monetary policy has some large unwanted side effects when it's tools are limited to interest rates & asset purchases. I can support that view but the answer is to give (credible) central banks more tools. I see scant historical evidence we can trust governments to control the money supply appropriately.

      Delete
    3. So your argument is: interest rate policy is effective because you believe it works.Well, that is how I described it, so not much I can say.

      Delete
    4. I don't believe it works, the market does. And the market weeds out bad thinkers over time. To me, the burden of evidence is always on who disagrees with the market (and, they should get very wealthy if correct).

      Don't we have a bit of a real-time experiment going on in Turkey. I believe Erdogan is citing Mosler-style theories when pushing for lower rates. It's not going well, to say the least.

      Delete
    5. I have no interest in Turkey. My point is simple: interest rate ineffectiveness us not the same thing as “interest rate have no effects on the economy.” Rather, it is that the effects are ambiguous, making it an uncertain policy lever. This the same thing as saying we do not know the effects of interest rate policy on the economy, which is a natural null hypothesis.

      If you disagree, with that, you are claiming to have knowledge of the effects of interest rate on the economy. Since you claim knowledge, the burden of proof is upon you. Think about it. If I used your logic, someone could say something that they are convinced that gargling saltwater cures cancer, and the burden of proof is to prove them wrong. That’s obviously not how science works.

      As for market reactions, that’s a non sequitur. Bond yields track the policy rate as a result of basic fixed income pricing. After that, market participants can make up any wacky theory they want for other asset classes.

      Delete
    6. On markets: of course rates follow the expected policy path - but inflation expectations need not. Spreads between inflation-linked and nominal bonds is market-determined.

      Agree the effect of interest rate policy is uncertain. The effect of all policy is uncertain. And if it's the government that wields such policy (as opposed to the central bank) i'd argue that the effects are even more uncertain.

      I'm not disagreeing with you so much as saying, if we want to change how conduct monetary/fiscal policy there is a fairly high burden of proof to show that a) we actually have a new plan, b) it is better, and c) governments can be trusted to apply it consistently.

      The relative silence of the MMT crowd in the face of high inflation indicates to me that they've not actually thought this through in any real-world sense. What policy levers do you pull to get inflation back on it's path? Can those levers actually be pulled? Do they get the current regime voted out of office if they pull them? If so, then what?

      Delete
    7. 1) Inflation-linked markets traders do not set the price of retail goods. I am not discussing financial markets, I am discussing the real economy,
      2) There’s a burden of proof that the existing “plan” works, which you are pretending does not exist because other people believe.
      3) Given the attachment of politicians to fiscal frameworks, there is mo reason to believe that they would ignore one that is explicitly aimed at inflation control.
      4) There is not a single credible person I am aware that does not believe that inflation is transitory. Your observation just tells us the obvious - very few hard money bugs are MMTers.

      Delete
    8. 《Inflation-linked markets traders do not set the price of retail goods. I am not discussing financial markets, I am discussing the real economy,》

      Why wouldn't real economy firms use inflation swaps and TIPS to manage their inflation risk?

      Delete
    9. Because they don’t have “inflation risk.” The risk they have is that input costs (including labour) rise faster than what they can charge.

      The problem with the inflation market is that the only real economy actor willing to sell inflation protection is the central government, and it is facing off against financial intermediaries that need to hedge inflation risk. Investment bankers have tried finding others, with no luck.

      Delete
  3. Great article. This is the sticky subject of course, and like you said it is somewhat of a standalone issue. You even have some people like Alex Howlett and Derek Gorder arguing that conventional monetary policy levers work more or less as presupposed.

    I am certainly of the persuasion of viewing rates as "out of equilibrium", and that loans are not so much about "creating" money ex-nihilo, but rather "upgrading" money from one form to another. So while conventional portfolio theory views a certain risk/return curve as being the "frontier" of an optimal portfolio, I think risk is not very quantifiable, at least once you start comparing across different asset classes, etc.

    But what is very quantifiable, is the "size" of an asset or investment. Like you said, housing ends up being where a lot of this plays out in "anglo" countries, it's a small enough asset that it's important to most consumers, but big enough that it nearly always requires financing. So in my view, the so called financial "intermediaries" end up playing a huge role in the setting of housing prices.

    But continuing with the theme of "non-equilibrium" rates and asset "size", I came up with a rough "schedule", that is to serve as an example of how rates might respond to "size". The equation for this "rate schedule" is r = 1/x^(1/3). Where r is the rate of growth as a decimal, and 1.00 corresponds to doubling every year. x is the "size" of the principal.

    So if you invest $1, the standard is that it doubles in value. But $1,000, has a rate of 0.1, or 10%. A million dollars would have a rate of 1%.

    This is really just a rough heuristic, to demonstrate a certain way in that rates might be "out of equilibrium". Once investors have enough money, it is not worth their time to dilly dally with small assets. For warren buffer it would not be worth his time to buy a local cupcake shop, even if the returns were high.

    Anyway, the idea that smaller things grow fast, also mirrors what we see in nature. It is not necessarily easy to quantify the growth rate of different organisms, but at least population growth rates are easily quantified and measure. But anyway, small organisms tend to grow faster.

    So this "size" story, offers an alternative, as to how a government might be able to secure lower rates, regardless of issues of operations, and money sovereignty.

    ReplyDelete
    Replies

    1. As for the effect of interest rates, I find perry mehrling's insights to be very instructive, especially understanding lender of last resort function, and how that played out historically. These lenders of last resort would often come in, to prevent a local economic collapse. Typically they could do this, because they represented a bigger geopolitical entity. In probability and statistics, aggregates simply diversify risk better. So with a larger geopolitical entity, historically, the risk between regions was very de-linked, and thus high powered government money might be more stable than local money.

      Well, the modern world, with consolidated supply chains and very global corporations, seems to work a little bit differently. So whereas a lender of last resort might have been there to alleviate local crises, today it seems like the entire system is continually dependent on high-powered money and the associated consolidated entities, like large corporations.

      Real resource seem to flow very much from the country to cities, in the modern world. Even with industrial centers, etc, purchasing power parity and cost of living difference, seem to indicate that there is a very real geographic flow of resources in one direction, ie prices are not symmetric.

      You could see this as cities creating money, and then the countryside needs that money to buy the industrial goods it relies on, but even so, cities have a lot of resources, and nominal incomes and GDP within cities are much higher.

      So the world seems to really have changed, in a way that would tend to support Mosler's and others reading of the situation. Rates are not a lender of last resort function, to patch up short spots, but a mechanism for controlling where resources are allocated and development occurs. This is because our dependence on high powered fiat is much higher than at any point historically. So raising rates, shapes the term structure, increases interest servicing costs for firms, and yet, it does little to impose balance sheet discipline on local banks. It may eat into their profits, but given that open market operations are just a perpetual standing thing, with large CB balance sheets.

      And of course, raising rates increases interest payments on government debt as well. Well, that's just some things to think about.

      I do think much of the pattern in housing prices is because of this dependence on high powered money, and like in 2008, housing has been one asset used to keep that high powered money flowing. So when housing prices increase, many balance sheets look better: bank balance sheets, city property taxes, even individual home owners. Well, we may have fixed the ability to use real estate to crash the financial system, I love hearing dylan ratigan talk about that, but it seems like we don't have any real price discipline.

      I know that Mosler want's to use asset side regulation for this discipline, I may not be the best party to judge whether this would be effective. Forcing banks to hold the loans they make may have this effect, I just am not familiar enough with banking to judge that. But personally, I think we should also consider our dependence on high powered money... If banks are not forced to solvent, but rather liquid and transparent, maybe they can act as "money creators", and maybe they would want low housing prices, which has positive externalities for communities. Just a thought.

      Delete
  4. This is great stuff, Brian. Really helpful for me around what Mosler was getting at about the interest rate channel adding money to the private sector, but also identifying how there are other effects.
    Where the housing market is concerned, Canadian housing starts peaked in 1976 -- that's 45 years ago!

    https://www.statista.com/statistics/198040/total-number-of-canadian-housing-starts-since-1995/

    I would propose that not least among the factors associated with the desperate housing situation in Canada is that we are governed for and by rentiers, but that takes us rather further into political science than is the subject of your blog, so I'll leave it at that.

    ReplyDelete
    Replies
    1. Thanks. The 1976 peak fits in with the Baby Boom, as the boomers needed housing.

      Delete

Note: Posts may be moderated, and there may be a considerable delay before they appear.

Although I welcome people who disagree with me, please be civil.

Please note that my spam comment filter appears to dislike long "anonymous" posts.

Note: if you want to post comments from Apple devices (iPhone, iPad), you apparently need to turn off "prevent cross-site tracking" in Safari privacy settings. (The reason presumably is that another URL handles comments, and so the user session needs to be preserved when redirected to that site. I don't like this, but this is not enough to make me switch my hosting service.)