I think the observation that 'time preference is very important' is correct. A low interest rate environment allowed all kinds of speculative ventures to take off, knowing they had years to borrow money cheaply and make money in the long term. That's changing now, and consequently any business that was doing that is suffering. The Metaverse thing Facebook has was a dubious proposition in a low rate environment and looks downright kamikaze now.
And it's great to see crypto suffer. Nobody has identified a use case for it except crooks. The only reason people bought it was to gamble that someone would buy it off you in the hope they could sell it to someone else for a higher price. That's a bad bet at anytime, but looks ridiculous at a time when you can get decent yield at low risk. Bitcoin mining is an environmental disaster.
"The only reason people bought it was to gamble that someone would buy it off you in the hope they could sell it to someone else for a higher price"
I know an unusually large number of people who are "into crypto". I don't think I can fairly describe the reasons behind their enthusiasm, at least not until I finally get a coffee this morning, but I can at least say it's nothing like greedy speculation - mostly it's idealist or even utopian ideas of how the financial system could be reworked so we don't have to rely on the "greedy corrupt bank heads and oligarchs" or something.
But speaking of speculation one guy I know did buy some amount of BTC in the very early days that's now worth some millions, but unfortunately for him he's never been able to remember his password.
Do the people you know ramble on about decentralisation and The Blockchain? Being your own bank is extremely hard. If you forget your password (!) you are completely screwed. Meanwhile, the mechanics of conventional banking are the best they have ever been. Smartphone banking apps are excellent. Paying money to others is pretty much instant and free or cheap. If you get defrauded you get your money back. I am also curious why people who want to be their own bank never want to be their own farmer, or butcher, or anything else.
I think there's a lot of pointing to the worst of the financial system's errors, say the run-up to the 2008 recession and too-big-to-fail banks being bailed out, and saying "let's do better!" then seeing bitcoin, etc. and thinking "power to the people! we don't need to trust Goldman Sach anymore".
Fwiw - all the crypto people I know I met at my previous job at a non-crypto fintech startup. So it's a very skewed sample. And the anti-elite attitude I described above only describes a few to varying degrees. Many of the rest are just excited about a new technology and think there are ways to use it and spread it around.
You could be right that criminal use cases are all crypto will ever amount to. I lean towards that opinion myself. But I can't dismiss the use cases some of these guys are trying to come up with. They'll probably fail just b/c 99% of new ideas fail, but who knows. All I can say is they don't think they're coming up with something that only criminals will use.
It's starting to remind me of fusion-based electricity generation. Yes, we've been working on it for decades, but the fantastic breakthrough is just around the corner. Yawn.
The main difference being, of course, that a successful fusion-based system would help save a warming planet and a successful proof of work crypto system would help fry it.
Does he have unlimited access to the wallet? Can he try passwords forever without consequence? If so, has he tried running actual cracking attempts against it? Certainly worth the attempt of a few million are in the line.
Good questions - he's definitely "tried everything" but I wonder if he's really tried the best cracking and hacking attempts. He needn't feel bad about it, since it's his,
But what I gather is emotionally he's moved on and prefers to just not think about it anymore. If this were text I'd add: lol, sigh, facepalm
I mean, there's very limited useful situations where crypto is all right, and it's basically limited to people in terrible countries being able to get sent funds or sending out funds, but outside of that, yup, gambling and crime.
Every use case for crypto is stuff I respond with, "I can do that more easily with my debit card right now."
It seems weird to have an entire article about tech woes and not mention at all the straightforward business decline being suffered by the big players in tech.
Part of the story here is that during the pandemic, ecommerce was up, online advertising was up, streaming services were up, and so forth. The companies bet that they could hang onto most of those gains in a post-pandemic world, and they were wrong. Now they have revenue problems.
That's not to say that time preference isn't also part of the story -- it's certainly part of the story that tech is telling itself -- but there's some pretty straightforward business decline too.
I think that’s why Matt referenced Derek Thompson’s Atlantic article, which addresses these themes, and then said he would take on the technical aspect as a sort of companion to the Thompson piece.
I just double-checked his article, and he doesn't say anything about business actually being down. He has all these vibes about casting around for innovation, but, for example, look at the graph in the Shopify layoff announcement:
"The story behind that long-term decline is itself interesting.."
This is probably a boring request, but I would love to see a write-up on that.
Bond yields & Treasuries are not sexy topics but have massive effects on economy / politics, and some insight on this relationship would be super interesting.
He has slightly updated versions of the paper but I don't think any are publicly available.
The short version is that it's mostly a combination of two things: the world has become safer (institutions, governments, policies, etc) so people require less interest to loan money out. You have a lot more faith that the modern UK government will repay you than you do that Henry VI would.
But there's also just been an increase in wealth globally, which translates into an increased supply of money, and increases in supply always reduce prices.
But wouldn’t that increase in wealth correspond with an increase in investment opportunities? Lots of ventures to sink money into now compared to 1500 or even 1900.
If you zoom out, then interest rates have been declining for centuries, or so I read. I can imagine several factors at play -- more efficient financial markets, abatement of extreme risks, global trade. The more recent four decade decline in 10 yr rate is said to be the result of a long period of moderate inflation. I’d like to see an analysis of both the centuries and the decades timescales.
At lot of it had to do with China keeping the yuan pegged to the USD during their manufacturing / export boom and then a global currency devaluation war after 2009.
Second this. The comment responses seem (at least on the surface) like they have some validity...just a general decline in risk in the world in general (less wars, better communications around the world, just more general knowledge) and better ability to calculate risk (modern computers and calculators make projecting risk much more precise).
One thing not mentioned is aging population. I think Krugman was where I read that one reason Japan's interest rates have remained low despite what was once thought to be an unmanageable debt load was an aging population. Suggests that while right now we are experiencing pretty sharp interest rate increases, in the medium to long term the better bet lower interest rates given American demographic trends (especially if our current anti-immigrant moment continues).
The paper I referenced above specifically disproves Krugman's claim that demographics are a driver of secular rate changes, though the details are relegated to an appendix rather than the main paper.
Per the article they went for 'get as many into people homes as possible by selling under cost and make money later' approach...and the 'make money later' bit never materialized.
I was given Alexas (Alexi?) alongside around six separate purchases of goods or services in 2020-21 and sold them all on eBay for an average of $25 per.
As a young-ish person, I for one would really like to see more speculative investments that I might be able to benefit from in a few decades and this newsletter really has me wondering what other alternatives we have to the Fed raising rates. Should we be relying on more fiscal policies to curb inflation, such as raising consumption taxes? Is it even possible for Congress and the President to effectively coordinate with the federal reserve if this is something they wanted to do? I'd definitely like to echo another commenter in saying I would love a history of how we found ourselves in the middle of a long term decline of the federal interest rate over the last four decades.
Just wait a while. Krugman says low interest rates will return. He may be right. Powell is trying to squeeze excess inflation out of the economy. Once that's done, demographic trends are likely to reassert themselves.
My understanding of the consumption tax proposals is the tax rate would by law automatically vary with the inflation rate. I'm sure there is a way to add to the laws governing the Federal Reserve that it take into account the consumption tax rate when determining the Funds Rate.
I think this is politically unfeasible. We currently have state governments handing out cash to "help people with inflation" and we want to do the opposite where things not only cost more because of inflation, but we're going to increase sales taxes on them too?
The consumption tax proposals swimming around aren't sales taxes. They are income taxes in which income is calculated through spending v. saving. Taxable income = gross income - standard deduction - savings. Raising the tax rate should incentivize saving and decrease aggregate demand.
It will work out to the same thing. Under inflationary conditions, people must spend more money in order to maintain the same standard of living. Raising taxes on people trying to maintain their standard of living is a political killer.
One of the reasons for the "tax revolt" in the late 70s and 80s in the US is that tax rates were not indexed to inflation and tax rates increased fairly dramatically from 1950 through 1980 due to inflation. This was especially true for the bottom half of income earners!
Yes, but raising interest rates kills investment, which people also don't like and can potentially trigger a recession. High inflation hurts, and stopping high inflation hurts. Tamping down consumer demand seems like the least worst option to me.
I'm just doubtful that its politically feasible. It might be like the carbon tax which is theoretically great, but so hated by the public as to be a political non starter.
Thank you as always Matt. I really like Thompson, but as a financial professional I found his dismissal of this case fairly odd. Fwiw - this is exactly what equity growth managers are saying on calls everyday to explain thier returns YTD.
Undeniably true, but Thompson is one of the writers I don't associate with those sort of narrative short cuts, but I'm not a consistent consumer of his work.
“Inflation was higher in 2021 and stayed high longer in 2022 than most forecasters predicted. And it’s important to remember that these were general errors. Joe Biden and his administration and their supporters made those mistakes. I made those mistakes. But so did Jerome Powell (a Republican), the Fed staff (a very nonpartisan group of professionals), and most private sector forecasters at the big investment banks. And crucially, the collective forecast of bond investors got it wrong.”
This true but I think misses the question as to why most everyone missed this. And generally why most experts fail to forecast major changes like this.
The two obvious reasons are:
- This is something that can’t be predicted.
- Groupthink and circular analysis.
Either way, it’s more evidence that long-term economic predictions from experts are not reliable, and this is what should be factored into future plans, not assumptions that the status quo ante will inevitably return.
I came here to reply that certainly not everyone got this wrong. It was a very frustrating 2021 watching our leadership accelerating towards the easily predictable cliff while they continued to get high on their own copium supply.
I think Matt and others should do deeper reflection on why they got it wrong and how they will prevent future mistakes. I agree with the assessment that it appears to be largely groupthink.
Had the #TeamTransitory circlejerk ended a few quarters sooner, the Fed wouldn’t have needed to slam the brakes so hard.
“When you’re surprised, you’re supposed to consider changing your mind” - Larry Summers
But beyond groupthink, I think there’s an issue of fighting the last war, and just not really wanting to accept that progressive economics can get it wrong, which is maybe also groupthink.
You're right that not everyone got this wrong, but I think it's an open question whether they were correct because they were lucky or correct because of better analysis. And if it's the latter, then what is the analysis and is it repeatable?
There are always a few people who correctly predict some big change whether it is inflation in 2021-2022, or the financial crisis in 2007-2008, but it's not often clear if they arrived at the correct conclusion via objective and repeatable analysis.
This is a bit off-topic, but relevant to Matt's many posts promoting zoning reform to encourage Yimbyism. Today's WSJ has an excellent article ('Land-Tax Policy Seen as Hurdle to Housing', p B6), which points out that many municipalities' tax structures tax buildings at higher rates than land, with the result that a lot of developable land under current zoning regimes remains either undeveloped or underdeveloped. In NYC, for instance, there are 77,000 lots that are either vacant or occupied by a building less than half the size permitted by current zoning rules: "If the owners of these lots all developed buildings with the maximum size allowed under current zoning, they could add an estimated 858 million square feet of housing and commercial space... Around 96% of these sites are zoned for residential use". The article argues for flipping the tax structure, so land is taxed at a higher rate than buildings (essentially Georgism). I haven't seen this addressed in any of Matt's pieces advocating Yimbyism, which all focus primarily on zoning reform. If the articles is correct, at least in NYC, that doesn't seem to be the primary constraint to housing development. Of course, rising interest rates crimp housing development too.
NYC doesn't even have to move to a LVT to fix this. They could just reduce property tax rates for commercial buildings and large rental apartment buildings to a reasonable number, like 1-3%.
Taxes on commercial and large rental buildings are so high that development only pencils out in NYC if the developer includes some affordable housing units so that they can take advantage of the state affordable housing tax abatement program (called 421a). This year the progressive state legislature declared this program a developer giveaway and declined to renew it.
Right around 3% on 10y rate I started seeing VC heads saying the money is gonna get shut off and you need to make profits now. It’s possible they were getting updates from end investors and passing that along to their startups, but higher interest rate->guidance change was very tightly coupled. The big tech companies seemed a few months slower, perhaps because their profit levels gave them more runway, but even this summer I was hearing intern horror stories about no one getting jobs.
I mean, that's true, but it's important to recognize the size of the shift, from "meh, who cares about profits, you can work that out in five years," to, "we need to be convinced that you have this plan on lock."
Well, it's a bad thing in a lot of ways, and probably a good thing in a lot of ways. But I'm not trying to say to make normalizing statements, I'm just saying that the magnitude of the shift is big. Yes, it doesn't go all the way to, "all companies must bootstrap, we will never provide any investment to companies that are not currently profitable," but it goes pretty far.
If the Fed ends up inventing successful time travel, then that'll have made inflation totally worth it. Probably safer than SERN's attempts, anyway.
More seriously, it's a bit odd to write a post on time discount rates and not mention the opportunity cost of locking up now-funds for later-interest. 20% interest on a $100 investment (security?) is a fantastic rate, but if I'm really poor and that $100 is a major chunk of my emergency reserve cash...well, who's to say what new crazy thing will happen within the year that might require sudden expenditures? Better to have and not need, and all that. (At an institutional level, I guess this would be reflected in cash reserves and such...which of course ties into FTX. Leverage is more dangerous in a less-stable economic situation. Dividends and profits now, please, as you say.)
ETA: also thank you for not mentioning the Stanford Marshmallow Prison Experiment.
The 40-year drop in interest rates probably explains a lot about the economy, honestly. Prior to tech taking center stage, we built a similar system in the financial industry. Rich people would shovel money into funds built by successful traders, seemingly oblivious to the fact that it’s basically just gambling.
I saw on Twitter last night that Matt complained about running low on ideas, and I thought to myself “where’s the post on interest rates I’ve been clamoring for?” Ask, and ye shall receive!
This is a great piece. As someone who works in a finance-adjacent field, I grok intuitively why rising rates have been having the impact on the economy that they have been, but haven’t really been able to articulate it well. The “bending time” frame is great and I will be using it.
No one knows where interest rates will be in the future, not even the fairly short-term future. MY writes "the 10-year rate is also rising, after having been in a period of structural decline for literally 40 years." But the accompanying chart shows frequent periods of similar increases, followed by declines. That's just as likely to happen now as not. Morningstar predicts Fed fund rates and 10-year Treasury yields in 2026 to be well below the historical average (https://www.morningstar.com/articles/1106505/why-we-expect-the-fed-to-cut-interest-rates-in-2023); Krugman writes that the era of cheap money may just be briefly interrupted, not ended (https://www.nytimes.com/2022/11/18/opinion/interest-rates-fed.html).
Anyone thinking the low interest-based economy we had has been left in the rearview mirror is making a pretty bold prediction. We just don't know yet.
“I do think a lot of us have the sense that the past 15 years have featured too much emphasis on throwing technical talent into super-duper-speculative ventures rather than something boring like “let’s make the self-checkout machines better” “
I don’t think it follows that now will be a better time to make these kinds of incremental innovations either. Whether it’s building Tesla or rolling out better check-out machines you are talking about deploying capital today at a loss in hopes of a future ROI. While projects with nearer payback and smaller total capital required may be better off on a relative basis than moonshots, the net effect is going to be making all innovation more expensive to finance. (I’m skeptical technical talent has been a major bottleneck on relatively simple incremental improvements.)
Companies working on low value stuff are miserly with engineering time. They micromanage you, decline to prioritize many bug fixes and improvements, and care a lot about deadlines, since it’s very easy to accidentally spend more labor on something than it’s worth. This tends to repel talent out to the Tech world where engineers are encouraged to be curious problem solvers, have and work on their own ideas, and just generally make things better however they can. Not to mention the wages are a lot higher.
If those alternatives disappear, you might get higher-caliber people working on lower-value stuff, but if they are smothered by cost-conscious management in the same way as the previous cohort, their results are unlikely to be much better.
I remember this dynamic playing out in the nineties with foreign investment in China (and emerging Asian markets generally).
In the first half of the decade people got very excited about including EM stocks in their portfolios. I was working for an investment bank in Hong Kong then and the business case seemed very solid.
But all it took was a modest rise in US rates (look at the 1994-1996 section of Matt's chart) to make debt-driven growth an unsustainable strategy for a lot of those companies, and even for some national economies.
"professionals with real money at stake were making decisions based on the idea that inflation would be much milder than it turned out to be"
And the TIPS market showing 5 and 10 year inflation expectations shows it. The Treasury really ought to give us more intermediate tenor TIPS to pin this down.
And the TIPS markets DID realize that inflation was let "transitory" in late Aug. 2021, months before the Fed raised the Federal Funds rates significantly. And very significantly, I think, those same markets now are expecting inflation over the nst 5 and 10 years to be spot on 2.3%, ~= the Fed's 2% PCE. target.
And I will continue to defend Biden economically from responsibility for the inflation,* except in the sense that it gave the Fed more "room" to make an upside error in. But ARA was not a secret that the Fed only discovered in December 2021. I will criticize Biden politically for a) claiming that ARA, etc. were "stimulus" rather than "relief" and b) not very vocally "having confidence: that the _Fed_ would make sure the the "temporary," supply side, Putin caused inflation was in fact temporary.
*You may or may not think that the size and destination of ARA and all the Trump relief spending was worth the growth inhibiting drag of its effect on the structural deficit.
Rather than relitigate yet again that both A) you can find out what inflation expectations are for the intermediate tenors by looking at secondary prices B) Treasury should absolutely not determine what securities to sell because hobbyists might find the data interesting occasionally. I think the fact that the long term rate is estimated to be at the Fed's target is just that because basically all bond traders have had basically their whole career in a secular bull market and can't imagine anything other than low inflation and rates are right around the corner.
I recently went through bond analyst calls from this year and in Feb (!) there were analysts calling that the bottom for the bond market. It's been really hard for people in the industry to shake the Pavlovian response that anytime rates go up that it's going to go even lower in the long-run.
If it's the case that intermediate expectations of inflation can be determined from secondary markets, then I'll modify my plea to asking FRED to publish them as they do 5 and 10 year rates. But who ever does it, the utility is not just to "hobbyists." :)
I'll take your point that expectations markets can be wrong, but they were at least directionally right about inflation sooner than the Fed was.
But why would should we think that interest rates will be much higher over the longer-term? Yes, they've gone up over the past year (and meanwhile all signs are pointing to declining inflation). To assume this is the new normal is a kind of availability bias.
I agree. This is a Krugman thing to try to predict real interest rates. It not necessary to run monetary policy.
It is relevant to whether the federal government should undertake certain long-lived projects. [Although to get super technical one needs to separate the capital scarcity cost from the time preference if some projects in the choice set have much longer lives than others.]
To be clear, I am also one of those bond analysts who has is experiencing the worst bond rout of their career and has been coping by saying things like "maybe next fall we'll be in a recession and rates will decline enough that all these projects will be feasible and we can make money again so hopefully we don't all get laid off in the interim." However, I do have concerns that I'm just saying that because rates have been in long-term decline my whole career (with shortish bouts of higher rates), and that I can't really comprehend a prolonged stretch of higher rates would be like.
I think there are good arguments for higher inflation for the short to medium term (deglobalization, reduced immigration, medium term increased costs associated with energy transitions, etc.). I do expect rates to go lower over the long-term. I just think there is a real risk that the market is underestimating the duration that moderately higher inflation may be with us.
I also think it's important to note that even if rates return to a more "normal" status that they'll be much higher than they were in 2020/2021. The ten year Treasury isn't going back to 0.60% unless there another huge intervention from the Fed that I just can't see happening anytime soon.
I think the observation that 'time preference is very important' is correct. A low interest rate environment allowed all kinds of speculative ventures to take off, knowing they had years to borrow money cheaply and make money in the long term. That's changing now, and consequently any business that was doing that is suffering. The Metaverse thing Facebook has was a dubious proposition in a low rate environment and looks downright kamikaze now.
And it's great to see crypto suffer. Nobody has identified a use case for it except crooks. The only reason people bought it was to gamble that someone would buy it off you in the hope they could sell it to someone else for a higher price. That's a bad bet at anytime, but looks ridiculous at a time when you can get decent yield at low risk. Bitcoin mining is an environmental disaster.
"The only reason people bought it was to gamble that someone would buy it off you in the hope they could sell it to someone else for a higher price"
I know an unusually large number of people who are "into crypto". I don't think I can fairly describe the reasons behind their enthusiasm, at least not until I finally get a coffee this morning, but I can at least say it's nothing like greedy speculation - mostly it's idealist or even utopian ideas of how the financial system could be reworked so we don't have to rely on the "greedy corrupt bank heads and oligarchs" or something.
But speaking of speculation one guy I know did buy some amount of BTC in the very early days that's now worth some millions, but unfortunately for him he's never been able to remember his password.
Do the people you know ramble on about decentralisation and The Blockchain? Being your own bank is extremely hard. If you forget your password (!) you are completely screwed. Meanwhile, the mechanics of conventional banking are the best they have ever been. Smartphone banking apps are excellent. Paying money to others is pretty much instant and free or cheap. If you get defrauded you get your money back. I am also curious why people who want to be their own bank never want to be their own farmer, or butcher, or anything else.
I think there's a lot of pointing to the worst of the financial system's errors, say the run-up to the 2008 recession and too-big-to-fail banks being bailed out, and saying "let's do better!" then seeing bitcoin, etc. and thinking "power to the people! we don't need to trust Goldman Sach anymore".
Fwiw - all the crypto people I know I met at my previous job at a non-crypto fintech startup. So it's a very skewed sample. And the anti-elite attitude I described above only describes a few to varying degrees. Many of the rest are just excited about a new technology and think there are ways to use it and spread it around.
You could be right that criminal use cases are all crypto will ever amount to. I lean towards that opinion myself. But I can't dismiss the use cases some of these guys are trying to come up with. They'll probably fail just b/c 99% of new ideas fail, but who knows. All I can say is they don't think they're coming up with something that only criminals will use.
It's starting to remind me of fusion-based electricity generation. Yes, we've been working on it for decades, but the fantastic breakthrough is just around the corner. Yawn.
The main difference being, of course, that a successful fusion-based system would help save a warming planet and a successful proof of work crypto system would help fry it.
Does he have unlimited access to the wallet? Can he try passwords forever without consequence? If so, has he tried running actual cracking attempts against it? Certainly worth the attempt of a few million are in the line.
Good questions - he's definitely "tried everything" but I wonder if he's really tried the best cracking and hacking attempts. He needn't feel bad about it, since it's his,
But what I gather is emotionally he's moved on and prefers to just not think about it anymore. If this were text I'd add: lol, sigh, facepalm
This would make me bug-chasing insane. I’d be a Twilight Zone protagonist.
I mean, there's very limited useful situations where crypto is all right, and it's basically limited to people in terrible countries being able to get sent funds or sending out funds, but outside of that, yup, gambling and crime.
Every use case for crypto is stuff I respond with, "I can do that more easily with my debit card right now."
It seems weird to have an entire article about tech woes and not mention at all the straightforward business decline being suffered by the big players in tech.
Part of the story here is that during the pandemic, ecommerce was up, online advertising was up, streaming services were up, and so forth. The companies bet that they could hang onto most of those gains in a post-pandemic world, and they were wrong. Now they have revenue problems.
That's not to say that time preference isn't also part of the story -- it's certainly part of the story that tech is telling itself -- but there's some pretty straightforward business decline too.
I think that’s why Matt referenced Derek Thompson’s Atlantic article, which addresses these themes, and then said he would take on the technical aspect as a sort of companion to the Thompson piece.
I just double-checked his article, and he doesn't say anything about business actually being down. He has all these vibes about casting around for innovation, but, for example, look at the graph in the Shopify layoff announcement:
https://news.shopify.com/changes-to-shopifys-team
That's a very stark graph, and it's not about either investment or innovation, it's about revenue declining.
"The story behind that long-term decline is itself interesting.."
This is probably a boring request, but I would love to see a write-up on that.
Bond yields & Treasuries are not sexy topics but have massive effects on economy / politics, and some insight on this relationship would be super interesting.
The story is told in Paul Schmelzing's "Eight centuries of global real interest rates, R-G, and the ‘suprasecular’ decline, 1311–2018"
https://www.bankofengland.co.uk/working-paper/2020/eight-centuries-of-global-real-interest-rates-r-g-and-the-suprasecular-decline-1311-2018
He has slightly updated versions of the paper but I don't think any are publicly available.
The short version is that it's mostly a combination of two things: the world has become safer (institutions, governments, policies, etc) so people require less interest to loan money out. You have a lot more faith that the modern UK government will repay you than you do that Henry VI would.
But there's also just been an increase in wealth globally, which translates into an increased supply of money, and increases in supply always reduce prices.
But wouldn’t that increase in wealth correspond with an increase in investment opportunities? Lots of ventures to sink money into now compared to 1500 or even 1900.
Matt wrote about this in the first month of Slow Boring!
https://www.slowboring.com/p/low-interest-rates-are-a-curse-we
If you zoom out, then interest rates have been declining for centuries, or so I read. I can imagine several factors at play -- more efficient financial markets, abatement of extreme risks, global trade. The more recent four decade decline in 10 yr rate is said to be the result of a long period of moderate inflation. I’d like to see an analysis of both the centuries and the decades timescales.
At lot of it had to do with China keeping the yuan pegged to the USD during their manufacturing / export boom and then a global currency devaluation war after 2009.
Second this. The comment responses seem (at least on the surface) like they have some validity...just a general decline in risk in the world in general (less wars, better communications around the world, just more general knowledge) and better ability to calculate risk (modern computers and calculators make projecting risk much more precise).
One thing not mentioned is aging population. I think Krugman was where I read that one reason Japan's interest rates have remained low despite what was once thought to be an unmanageable debt load was an aging population. Suggests that while right now we are experiencing pretty sharp interest rate increases, in the medium to long term the better bet lower interest rates given American demographic trends (especially if our current anti-immigrant moment continues).
The paper I referenced above specifically disproves Krugman's claim that demographics are a driver of secular rate changes, though the details are relegated to an appendix rather than the main paper.
Re: big swings, Amazon's Alexa is going to end the year with something like a $10B loss, probably with no end in sight other than, well, ending it:
https://arstechnica.com/gadgets/2022/11/amazon-alexa-is-a-colossal-failure-on-pace-to-lose-10-billion-this-year/
I like my Alexa. I definitely would have payed more than $35 for it too. Why didn't Amazon even attempt to sell them for profit?
Per the article they went for 'get as many into people homes as possible by selling under cost and make money later' approach...and the 'make money later' bit never materialized.
Just a bad bet.
I paid $0.00 for mine, which seems about right.
I was given Alexas (Alexi?) alongside around six separate purchases of goods or services in 2020-21 and sold them all on eBay for an average of $25 per.
Lol.
As a young-ish person, I for one would really like to see more speculative investments that I might be able to benefit from in a few decades and this newsletter really has me wondering what other alternatives we have to the Fed raising rates. Should we be relying on more fiscal policies to curb inflation, such as raising consumption taxes? Is it even possible for Congress and the President to effectively coordinate with the federal reserve if this is something they wanted to do? I'd definitely like to echo another commenter in saying I would love a history of how we found ourselves in the middle of a long term decline of the federal interest rate over the last four decades.
Just wait a while. Krugman says low interest rates will return. He may be right. Powell is trying to squeeze excess inflation out of the economy. Once that's done, demographic trends are likely to reassert themselves.
Here's an archived snippet of his thinking on this: https://archive.ph/shkWt
Matt wrote about this in 2020: https://www.slowboring.com/p/low-interest-rates-are-a-curse-we
My understanding of the consumption tax proposals is the tax rate would by law automatically vary with the inflation rate. I'm sure there is a way to add to the laws governing the Federal Reserve that it take into account the consumption tax rate when determining the Funds Rate.
I think this is politically unfeasible. We currently have state governments handing out cash to "help people with inflation" and we want to do the opposite where things not only cost more because of inflation, but we're going to increase sales taxes on them too?
The consumption tax proposals swimming around aren't sales taxes. They are income taxes in which income is calculated through spending v. saving. Taxable income = gross income - standard deduction - savings. Raising the tax rate should incentivize saving and decrease aggregate demand.
It will work out to the same thing. Under inflationary conditions, people must spend more money in order to maintain the same standard of living. Raising taxes on people trying to maintain their standard of living is a political killer.
One of the reasons for the "tax revolt" in the late 70s and 80s in the US is that tax rates were not indexed to inflation and tax rates increased fairly dramatically from 1950 through 1980 due to inflation. This was especially true for the bottom half of income earners!
Yes, but raising interest rates kills investment, which people also don't like and can potentially trigger a recession. High inflation hurts, and stopping high inflation hurts. Tamping down consumer demand seems like the least worst option to me.
It might be the least worst option!
I'm just doubtful that its politically feasible. It might be like the carbon tax which is theoretically great, but so hated by the public as to be a political non starter.
Thank you as always Matt. I really like Thompson, but as a financial professional I found his dismissal of this case fairly odd. Fwiw - this is exactly what equity growth managers are saying on calls everyday to explain thier returns YTD.
I just think he found the vibes story more interesting, so he has to discard the economic story to get back to his preferred narrative.
It's unfortunately a common feature writing move.
Undeniably true, but Thompson is one of the writers I don't associate with those sort of narrative short cuts, but I'm not a consistent consumer of his work.
“Inflation was higher in 2021 and stayed high longer in 2022 than most forecasters predicted. And it’s important to remember that these were general errors. Joe Biden and his administration and their supporters made those mistakes. I made those mistakes. But so did Jerome Powell (a Republican), the Fed staff (a very nonpartisan group of professionals), and most private sector forecasters at the big investment banks. And crucially, the collective forecast of bond investors got it wrong.”
This true but I think misses the question as to why most everyone missed this. And generally why most experts fail to forecast major changes like this.
The two obvious reasons are:
- This is something that can’t be predicted.
- Groupthink and circular analysis.
Either way, it’s more evidence that long-term economic predictions from experts are not reliable, and this is what should be factored into future plans, not assumptions that the status quo ante will inevitably return.
I came here to reply that certainly not everyone got this wrong. It was a very frustrating 2021 watching our leadership accelerating towards the easily predictable cliff while they continued to get high on their own copium supply.
I think Matt and others should do deeper reflection on why they got it wrong and how they will prevent future mistakes. I agree with the assessment that it appears to be largely groupthink.
Had the #TeamTransitory circlejerk ended a few quarters sooner, the Fed wouldn’t have needed to slam the brakes so hard.
https://www.cnn.com/2021/05/26/economy/inflation-larry-summers-biden-fed/index.html
“When you’re surprised, you’re supposed to consider changing your mind” - Larry Summers
But beyond groupthink, I think there’s an issue of fighting the last war, and just not really wanting to accept that progressive economics can get it wrong, which is maybe also groupthink.
https://auth.jacobinmag.com/2021/09/larry-summers-inflation-predictions-biden-stimulus-american-rescue-plan-federal-reserve-treasury
You're right that not everyone got this wrong, but I think it's an open question whether they were correct because they were lucky or correct because of better analysis. And if it's the latter, then what is the analysis and is it repeatable?
There are always a few people who correctly predict some big change whether it is inflation in 2021-2022, or the financial crisis in 2007-2008, but it's not often clear if they arrived at the correct conclusion via objective and repeatable analysis.
This is a bit off-topic, but relevant to Matt's many posts promoting zoning reform to encourage Yimbyism. Today's WSJ has an excellent article ('Land-Tax Policy Seen as Hurdle to Housing', p B6), which points out that many municipalities' tax structures tax buildings at higher rates than land, with the result that a lot of developable land under current zoning regimes remains either undeveloped or underdeveloped. In NYC, for instance, there are 77,000 lots that are either vacant or occupied by a building less than half the size permitted by current zoning rules: "If the owners of these lots all developed buildings with the maximum size allowed under current zoning, they could add an estimated 858 million square feet of housing and commercial space... Around 96% of these sites are zoned for residential use". The article argues for flipping the tax structure, so land is taxed at a higher rate than buildings (essentially Georgism). I haven't seen this addressed in any of Matt's pieces advocating Yimbyism, which all focus primarily on zoning reform. If the articles is correct, at least in NYC, that doesn't seem to be the primary constraint to housing development. Of course, rising interest rates crimp housing development too.
NYC doesn't even have to move to a LVT to fix this. They could just reduce property tax rates for commercial buildings and large rental apartment buildings to a reasonable number, like 1-3%.
Large apartment buildings (Class 2) are currently taxed at about 6% (12.2% of the assessed value, which is 45% of the estimated market value: https://www.nyc.gov/site/finance/taxes/property-tax-rates.page).
Taxes on commercial and large rental buildings are so high that development only pencils out in NYC if the developer includes some affordable housing units so that they can take advantage of the state affordable housing tax abatement program (called 421a). This year the progressive state legislature declared this program a developer giveaway and declined to renew it.
Right around 3% on 10y rate I started seeing VC heads saying the money is gonna get shut off and you need to make profits now. It’s possible they were getting updates from end investors and passing that along to their startups, but higher interest rate->guidance change was very tightly coupled. The big tech companies seemed a few months slower, perhaps because their profit levels gave them more runway, but even this summer I was hearing intern horror stories about no one getting jobs.
I mean, that's true, but it's important to recognize the size of the shift, from "meh, who cares about profits, you can work that out in five years," to, "we need to be convinced that you have this plan on lock."
Well, it's a bad thing in a lot of ways, and probably a good thing in a lot of ways. But I'm not trying to say to make normalizing statements, I'm just saying that the magnitude of the shift is big. Yes, it doesn't go all the way to, "all companies must bootstrap, we will never provide any investment to companies that are not currently profitable," but it goes pretty far.
If the Fed ends up inventing successful time travel, then that'll have made inflation totally worth it. Probably safer than SERN's attempts, anyway.
More seriously, it's a bit odd to write a post on time discount rates and not mention the opportunity cost of locking up now-funds for later-interest. 20% interest on a $100 investment (security?) is a fantastic rate, but if I'm really poor and that $100 is a major chunk of my emergency reserve cash...well, who's to say what new crazy thing will happen within the year that might require sudden expenditures? Better to have and not need, and all that. (At an institutional level, I guess this would be reflected in cash reserves and such...which of course ties into FTX. Leverage is more dangerous in a less-stable economic situation. Dividends and profits now, please, as you say.)
ETA: also thank you for not mentioning the Stanford Marshmallow Prison Experiment.
The 40-year drop in interest rates probably explains a lot about the economy, honestly. Prior to tech taking center stage, we built a similar system in the financial industry. Rich people would shovel money into funds built by successful traders, seemingly oblivious to the fact that it’s basically just gambling.
I saw on Twitter last night that Matt complained about running low on ideas, and I thought to myself “where’s the post on interest rates I’ve been clamoring for?” Ask, and ye shall receive!
This is a great piece. As someone who works in a finance-adjacent field, I grok intuitively why rising rates have been having the impact on the economy that they have been, but haven’t really been able to articulate it well. The “bending time” frame is great and I will be using it.
No one knows where interest rates will be in the future, not even the fairly short-term future. MY writes "the 10-year rate is also rising, after having been in a period of structural decline for literally 40 years." But the accompanying chart shows frequent periods of similar increases, followed by declines. That's just as likely to happen now as not. Morningstar predicts Fed fund rates and 10-year Treasury yields in 2026 to be well below the historical average (https://www.morningstar.com/articles/1106505/why-we-expect-the-fed-to-cut-interest-rates-in-2023); Krugman writes that the era of cheap money may just be briefly interrupted, not ended (https://www.nytimes.com/2022/11/18/opinion/interest-rates-fed.html).
Anyone thinking the low interest-based economy we had has been left in the rearview mirror is making a pretty bold prediction. We just don't know yet.
“I do think a lot of us have the sense that the past 15 years have featured too much emphasis on throwing technical talent into super-duper-speculative ventures rather than something boring like “let’s make the self-checkout machines better” “
I don’t think it follows that now will be a better time to make these kinds of incremental innovations either. Whether it’s building Tesla or rolling out better check-out machines you are talking about deploying capital today at a loss in hopes of a future ROI. While projects with nearer payback and smaller total capital required may be better off on a relative basis than moonshots, the net effect is going to be making all innovation more expensive to finance. (I’m skeptical technical talent has been a major bottleneck on relatively simple incremental improvements.)
Companies working on low value stuff are miserly with engineering time. They micromanage you, decline to prioritize many bug fixes and improvements, and care a lot about deadlines, since it’s very easy to accidentally spend more labor on something than it’s worth. This tends to repel talent out to the Tech world where engineers are encouraged to be curious problem solvers, have and work on their own ideas, and just generally make things better however they can. Not to mention the wages are a lot higher.
If those alternatives disappear, you might get higher-caliber people working on lower-value stuff, but if they are smothered by cost-conscious management in the same way as the previous cohort, their results are unlikely to be much better.
I remember this dynamic playing out in the nineties with foreign investment in China (and emerging Asian markets generally).
In the first half of the decade people got very excited about including EM stocks in their portfolios. I was working for an investment bank in Hong Kong then and the business case seemed very solid.
But all it took was a modest rise in US rates (look at the 1994-1996 section of Matt's chart) to make debt-driven growth an unsustainable strategy for a lot of those companies, and even for some national economies.
"professionals with real money at stake were making decisions based on the idea that inflation would be much milder than it turned out to be"
And the TIPS market showing 5 and 10 year inflation expectations shows it. The Treasury really ought to give us more intermediate tenor TIPS to pin this down.
And the TIPS markets DID realize that inflation was let "transitory" in late Aug. 2021, months before the Fed raised the Federal Funds rates significantly. And very significantly, I think, those same markets now are expecting inflation over the nst 5 and 10 years to be spot on 2.3%, ~= the Fed's 2% PCE. target.
And I will continue to defend Biden economically from responsibility for the inflation,* except in the sense that it gave the Fed more "room" to make an upside error in. But ARA was not a secret that the Fed only discovered in December 2021. I will criticize Biden politically for a) claiming that ARA, etc. were "stimulus" rather than "relief" and b) not very vocally "having confidence: that the _Fed_ would make sure the the "temporary," supply side, Putin caused inflation was in fact temporary.
*You may or may not think that the size and destination of ARA and all the Trump relief spending was worth the growth inhibiting drag of its effect on the structural deficit.
Rather than relitigate yet again that both A) you can find out what inflation expectations are for the intermediate tenors by looking at secondary prices B) Treasury should absolutely not determine what securities to sell because hobbyists might find the data interesting occasionally. I think the fact that the long term rate is estimated to be at the Fed's target is just that because basically all bond traders have had basically their whole career in a secular bull market and can't imagine anything other than low inflation and rates are right around the corner.
I recently went through bond analyst calls from this year and in Feb (!) there were analysts calling that the bottom for the bond market. It's been really hard for people in the industry to shake the Pavlovian response that anytime rates go up that it's going to go even lower in the long-run.
If it's the case that intermediate expectations of inflation can be determined from secondary markets, then I'll modify my plea to asking FRED to publish them as they do 5 and 10 year rates. But who ever does it, the utility is not just to "hobbyists." :)
I'll take your point that expectations markets can be wrong, but they were at least directionally right about inflation sooner than the Fed was.
But why would should we think that interest rates will be much higher over the longer-term? Yes, they've gone up over the past year (and meanwhile all signs are pointing to declining inflation). To assume this is the new normal is a kind of availability bias.
I agree. This is a Krugman thing to try to predict real interest rates. It not necessary to run monetary policy.
It is relevant to whether the federal government should undertake certain long-lived projects. [Although to get super technical one needs to separate the capital scarcity cost from the time preference if some projects in the choice set have much longer lives than others.]
To be clear, I am also one of those bond analysts who has is experiencing the worst bond rout of their career and has been coping by saying things like "maybe next fall we'll be in a recession and rates will decline enough that all these projects will be feasible and we can make money again so hopefully we don't all get laid off in the interim." However, I do have concerns that I'm just saying that because rates have been in long-term decline my whole career (with shortish bouts of higher rates), and that I can't really comprehend a prolonged stretch of higher rates would be like.
I think there are good arguments for higher inflation for the short to medium term (deglobalization, reduced immigration, medium term increased costs associated with energy transitions, etc.). I do expect rates to go lower over the long-term. I just think there is a real risk that the market is underestimating the duration that moderately higher inflation may be with us.
I also think it's important to note that even if rates return to a more "normal" status that they'll be much higher than they were in 2020/2021. The ten year Treasury isn't going back to 0.60% unless there another huge intervention from the Fed that I just can't see happening anytime soon.