The standard line from Modern Monetary Theory (MMT) on interest rates has long been to advocate for a low-and-steady or set-it-and-forget-it approach: set the overnight risk-free interest rate low, like maybe zero, and then just leave it there. While I used to be completely onboard with this approach, I have lately started to rethink it slightly. In this article I’m going to explain why.
The MMT Approach to Interest Rate Policy
To start, why do MMTers like the set-it-and-forget-it strategy ?There are a number of arguments that align to produce this recommendation.
One is a rejection of the ability of the monetary authorities to “fine-tune” the economy. Contrary to standard narrative, MMTers claim, the effects of interest rate changes on the economy are complex and variable, so it’s not as simple as ‘lower rates = gas, higher rates = brake.’ There are numerous effects which push in the opposite direction, the magnitudes of which depend on the state of the economy. These include interest costs of financing inventory, rate effects on forward prices, consumption behavior by target-savers, target rate of return pricing, and income from the interest expenditure on government debt. Furthermore, an increasing amount of production in countries like the US takes place in markets that aren’t very sensitive to interest rates. So, exactly what the outcome of raising or lowering rates on demand and inflation will be is hard to say. Therefore, we say, stabilization is better left to fiscal policy (preferably through automatic stabilizers like the Job Guarantee).
Another argument is about distribution. Lower rates are more attractive in that they don’t provide unearned income to rentiers, and make it easier to start new businesses, invest in innovation, and so forth. As Warren Mosler likes to say, a positive risk-free rate is a basic income to people who already have money.
Finally, another line of reasoning concerns financial stability. This draws on Hyman Minsky’s work on financial fragility, and particularly his famous taxonomy of Hedge, Speculative, and Ponzi financial positions. As a quick refresher, a Hedge borrower is one whose expected income from an investment is more than enough to cover both the principal and interest on the loans funding that investment. A Speculative borrower is one for whom the expected income can cover the interest but not the principal payments, meaning that when the loan comes due, the borrower will need to refinance, rolling the loan over. A Ponzi borrower is one whose expected income isn’t even high enough to cover the interest payments, meaning the Ponzi unit has to continually borrow more just to service the loan. Both Speculative and Ponzi agents depend on access to financial markets to make their debt payments, and are therefore more fragile than Hedge units — and an economy dominated by Speculative and Ponzi borrowers will be more fragile than one dominated by Hedge borrowers. The key point is that increases in debt service costs, like when interest rates rise, will increase financial fragility, pushing units from Hedge toward Speculative and Ponzi and then default.
This dovetails with the first argument too: raising rates doesn’t work like a surgical tool that reduces demand a predictable amount — rather, the effect on demand is uncertain up to the point where rates are so high that they break the financial system, causing a financial crisis that then cuts off refinancing and new funding. Not a civilized way to manage an economy.
Now, there is one line of thinking, outside of MMT, that says that we should raise rates with the intention of causing something like a financial crisis, not just anytime or willy-nilly, but as a way of popping bubbles. For example, some argue that if the Fed had raised rates more aggressively earlier, then they could have averted the damage from the housing bubble or the dotcom bubble, by just popping these bubbles before they got too big. I’d say this argument leans anywhere from centrist-technocratic to right-libertarian (where the latter would go farther and say the Fed’s low rates caused the bubble in the first place).
The MMT response has generally been that this is not the right way to deal with bubbles. Bubbles aren’t just about “irrational exuberance,” they’re often about widespread fraud, and to deal with these criminals we need better police. Specifically we need strong and flexible regulators, who can keep pace on developments in financial markets and even anticipate new directions that will innovate around existing rules.
And I agree with all of this. But.
Reconsidering Bubbles
I have to say, I have been struck by the unreasonable efficiency by which the Fed raising rates just a few points led to a widespread collapse of cryptocurrency ponzi schemes in 2022. The crypto sector has been frothing with fraud for quite some time, and while regulators are slowly moving in, it seems like quite a lot of their work was done for them by the Fed in just a few months. The Fed raised rates, causing the prices of risky assets to drop as savers re-allocated toward safer assets that now pay higher yields. This hurt the ponzis on both fronts, tightening inflows while broadening outflows, and many of them promptly collapsed. See FTX, etc.
This has made me reconsider my position on interest rates.
I think we can look at it this way. Broadly speaking we’ve got two categories of policy tools: one set that are very blunt and impact everything across the economy, and another set that are surgical and impact only specific firms or sectors. In the “blunt” category go interest rate adjustments, while in the “surgical” category go things like specific credit controls, regulation, industrial policy, subsidies, and the like.
In this framing, the MMT argument has been that interest rates are too blunt of a tool, so we shouldn’t use them for anything. Yes they can pop bubbles, but this causes too much collateral damage to legitimate financial and investment activities. We should instead use our surgical tools like regulation to stop the fraudulent activity.
So I think I’ve got two objections to this approach: one is about speed, the other about comprehensiveness.
On speed: basically as a matter of logic, surgical tools take more time to work. That’s because you can’t use the surgical tool until it’s been adapted to the job at hand. New regulations have to be crafted for the particular scenario we find ourselves in, which of course doesn’t even begin to happen until we’re widely cognizant that new developments may be a threat, and won’t be completed until some time later. I do think we could speed this up significantly compared to what we do now, but some lag is unavoidable. Then of course there’s the lag where enforcement catches up with the new rules.
On comprehensiveness: the issue here is that the more surgical your tool is, the more likely it is for crumbs to slip through the cracks. What I mean is, regulators can only come down on violations that are visible to regulators, which usually means organized business. They probably can’t crack down on, say, your cousin Mike running a small-time Ponzi scheme with friends and family — they just have no way of knowing about that, at least not without us dramatically increasing surveillance beyond what reasonable people want (already a huge problem).
I’m suggesting we rethink the idea that the surgical tool does everything and the blunt tool does nothing. Instead, I think we should 1) use the surgical tools most of the time, 2) keep the blunt tools in our back pocket to be used occasionally as needed, and 3) also use the surgical tools to counteract the the damaging effects of the blunt tools.
To use another analogy, what do you do when you see a bug in your home? You kill it (or if you’re like me, throw it outside). What if you see 10 bugs? You go kill them one-by-one (or throw each one outside!).
What if you see 1,000 bugs? You don’t go after them one-by-one. You fumigate. If killing bugs one-by-one is the surgical approach, fumigating the house is the blunt instrument. It’s time to fumigate when you recognize that the bugs are everywhere, so a) the time/effort of going after them one-by-one isn’t worth it, and b) many would slip through the cracks (literally) if you went after them one-by-one. Fumigating gets them all, everywhere.
Fumigating also does substantial collateral damage. But this isn’t an argument to never fumigate — rather we use other tools to alleviate that damage. These other tools are in fact surgical, eg. removing particular items that might be damaged by fumes, finding temporary locations for each person or pet to stay, and so on.
In concrete terms, I think that means the following. Most of the time we should go by the standard MMT approach of keeping interest rates low and steady, to promote legitimate investment. But we should keep interest rate hikes in our back pocket as something that can be done when it seems like the one-by-one approach isn’t going to cut it, i.e. once the bubble has already gotten out of hand. Raising rates lets us collapse ponzi schemes that we don’t even know about.
In other words, yes, I believe that prevention is the best medicine, but we should also recognize that prevention doesn’t always work, and that the prophylactic treatment and the therapeutic treatment might not be the same treatment! Since I’m throwing all kinds of metaphors around, here’s one more: most of the time financial activity can be considered innocent-until-proven-guilty, but every now and again this flips and that activity should be considered guilty-until-proven-innocent, and then our approach needs to change.
Managing the Fallout
But of course raising rates causes substantial collateral damage if not mitigated, so we need to employ some of our other surgical tools to ameliorate this. We can do that by specifically shielding key sectors and institutions from the effects of higher rates. The government is already involved in steering credit in all kinds of ways so this isn’t that new. (For instance, how about when the government insures loans to farmers? Or when government-sponsored enterprises created the mortgage-backed security market? Or when the federal government holds most of the student debt in this country? You get the idea.) What would be different is that we would be increasing support for key areas at the same time that we otherwise tighten conditions across the board —pro-cyclical credit support to counteract our counter-cyclical interest rate policy.
Now, I don’t claim to know exactly what this additional support should look like. This is something that will require a lot of careful thought. What we’d be trying to do essentially is decrease certain credit spreads at the same time that we raise Treasury rates. Generally the way we reduce certain spreads is through insuring private loans, increasing secondary market liquidity, subsidizing an asset through tax exemptions, or just having the government make loans at below-market rates. I’m kind of thinking that the easiest way to go is to pay people for holding certain assets. This should reduce interest rates for borrowers issuing those liabilities. So, the government could simultaneously raise the Fed Funds rate and raise the amount it pays to holders of a target class of assets.
There then may be some concern that we’re subsidizing savers too much. On the whole, I would agree, although there is maybe a small argument to be made that they’re working in the public interest when we recruit them to help pop bubbles. But, if we’re worried about it, we can link the whole scheme to an overall wealth tax: a wealth tax would claw back the subsidy to savers, but shouldn’t affect spreads.
Final Thoughts
So then the formula would be: simultaneously raise interest rates, increase support to desired sectors, and increase the wealth tax rate. Decrease each of these when the episode is over.
One last point. Unlike bugs who are caught by surprise by fumigation, financial fraudsters could see higher rates coming, and prepare so as to weather the storm. At the risk of stretching the analogy, we don’t just have bugs, we also have super-bugs, who can predict that fumigation is about to happen and so leave the house, to return after the tent is gone. To cope with this, we probably need to turn the fumigation (interest rate hikes) once again into something of a surprise. This could be difficult — the super-bugs may be smart enough to eventually learn the telltale signs that you’re about to fumigate. There may actually be no solution to this — however, even if fumigating doesn’t kill all the super-bugs, it will at least get the regular ones. And presumably the super-bugs will usually be fewer in number than the regular bugs, and so more amenable to our use of the surgical approach.
I will acknowledge that this is getting a bit rickety and ad hoc… But remember that by design this is something of an emergency measure, not day-to-day policy. I’m not saying we should use interest rates to fine-tune for financial stability, only that raising rates to wipe out pests isn’t something that we should promise in advance to never do.