Inflation & Interest Rates: share your data sources, models, and assumptions

[reeshau]

The WSJ has an article today on the difficulty in estimating household savings, resulting in a wide variety of measures.  The difficulty is not just measuring what's going on now, but also the variability of pre-pandemic data to use as a base rate.  For example, the Commerce Department initially reported that the saving rate in 2012 was 3.9%, for example. Now the 2012 reading is 8.6%.

Bank of America, measuring account balances directly, shows people of all incomes still with 150% of pre-pandemic cash.  I would think some of this is money that had been in risk accounts, which has come back to banks with higher interest rates.  But BoA is still only paying 0.1% on its money market savings, so that may not be a factor in their numbers.

[EscapeVelocity2020]

The WSJ has an article today on the difficulty in estimating household savings, resulting in a wide variety of measures.  The difficulty is not just measuring what's going on now, but also the variability of pre-pandemic data to use as a base rate.  For example, the Commerce Department initially reported that the saving rate in 2012 was 3.9%, for example. Now the 2012 reading is 8.6%.

Bank of America, measuring account balances directly, shows people of all incomes still with 150% of pre-pandemic cash.  I would think some of this is money that had been in risk accounts, which has come back to banks with higher interest rates.  But BoA is still only paying 0.1% on its money market savings, so that may not be a factor in their numbers.

Unfortunately the article is behind a paywall, but I'm sure a non-insignificant amount of bank savings has moved to money market accounts paying 4 - 5%.  Not sure how easy it is to track that since money market accounts (and other high yield online 'saving-ish' accounts) blur the line between traditional bank savings and financial institution investments. 

[reeshau]

That's what made the source interesting to me.  B of A isn't paying anything near that, on a demand account anyway.

[ChpBstrd]

August PCE:           +0.4%       3.5% annualized
August Core PCE:   +0.1%       3.9% annualized

Real personal income excluding transfer receipts were up +0.1%, but real disposable income was down -.2%. Over the past 12 months, disposable income has risen +3.7% and this has been slowing rapidly since June, when it was +5.4%. So much for the narrative of wage inflation. It seems to be disappearing quickly, though I'm unsure how to reconcile these numbers with the rising wage trend and the flat-ish hours trend.

Edit: Also interesting is how the BEA is now excluding housing in a new version of Core PCE. According to Reuters:

With the August data, the government introduced new price measures, the PCE price index excluding food, energy and housing, and PCE services excluding energy and housing, the so-called super core inflation.

The PCE price index excluding food, energy and housing also gained 0.1% last month after rising 0.2% in July. PCE services excluding energy and housing inflation rose 0.1% after climbing 0.5% in the prior month.

[MustacheAndaHalf]

I prefer this interactive page for wage growth data, which lets me compare 12 month average to 3 month average for certain categories.  For job switcher vs job stayer, you can see the reward of switching jobs.
https://www.atlantafed.org/chcs/wage-growth-tracker

12 month average was +6.8% switching vs +5.5% staying
3 month average was +5.6% switching vs +5.2% staying

I don't know if summer vacations influenced those numbers.  It looks like recently, switching jobs isn't rewarded by much.  The news coverage of striking workers may be dramatic, but union workers are a small fraction of overall workers.  The trend I see is employers not paying as much for new employees, which suggests workers may be losing their ability to pressure employers.

As a consistency check, I looked at the unemployment rate.  It spiked in March 2022, after which it kept falling until it moved up slightly in Jan 2022.  During 2022, unemployment fell from 4.0% to 3.5%.  In 2023 unemployment has been bouncing around a lot, so I think tracking the peaks and valleys separately can smooth out the bumps.
https://fred.stlouisfed.org/series/UNRATE

2023 Peaks: 3.6% 3.7% 3.8%
2023 Lows: 3.4% 3.4% 3.5%

It looks like unemployment is slowly moving upwards, which matches what I see in job switchers vs job stayers.

[ChpBstrd]

Here's a very interesting analysis by the Money & Macro channel, which looks at how current research is attempting to break down the causes of inflation (supply disruptions by COVID and Russia, or demand spikes from stimulus).

https://www.youtube.com/watch?v=7GMSzNm6i_U

TL;DW:

• About 60% of inflation in Europe can be attributed to supply disruptions from the Ukraine war, 20% to COVID, and 20% to other reasons.
• In the U.S, core inflation explained most of the increases, with housing/rents explaining a disproportionate amount of inflation compared to Europe. Demand seems to be a bigger culprit in the U.S.
• "Team Transitory" was right in the sense that one-off factors explained inflation, but the timeframe was maybe off.

[MustacheAndaHalf]

The original "team transitory" was in mid-2021 to late-2021, who claimed that inflation was going to fall back to zero on its own.  That didn't happen, and the Fed admitted "transitory" was wrong in late 2021.  Despite that, people talked of transitory again in early 2022... and inflation spiked higher, rather than going away in 6 months.

If someone guesses inflation will go away in 6 months, they may eventually be correct.  But it helps to look back at "team transitory" track record, which is terrible.

[ChpBstrd]

The original "team transitory" was in mid-2021 to late-2021, who claimed that inflation was going to fall back to zero on its own.  That didn't happen, and the Fed admitted "transitory" was wrong in late 2021.  Despite that, people talked of transitory again in early 2022... and inflation spiked higher, rather than going away in 6 months.

If someone guesses inflation will go away in 6 months, they may eventually be correct.  But it helps to look back at "team transitory" track record, which is terrible.

Transitory in 6 months was a failed call, but what if they had said "in 12-24 months we'll see inflation plummeting due to a restoration of supply chains"? Well, that kinda happened too, starting in June 2022. I suspect the timeframe was supplied by the media, rather than the economists. It's also worth noting how some of the economists and influencers who were right in June 2021 about the inflation risk were wrong in June 2022 about the onset of disinflation.

The studies cited in the video are an interesting attempt to tease apart the pandemic supply constraints versus the stimulus-induced boom in demand - both one-time events that self-resolved. The explanation matters because the US had much more stimulus than the EU, and yet the EU had a similar boom in inflation, especially around the Ukraine war. So was it supply? The research suggests different explanations for the two economies' near-simultaneous experiences of inflation.

Most of the factors affecting inflation in the US are in the rear-view mirror (supply chains and stimulus), and resuming student loan payments will resemble an anti-stimulus or even more QT. We really need to know whether inflation is due to money supply remaining far above pre-pandemic levels or if we just went through a stretch of one-time supply and demand events. The Fed seems to be expecting a multi-year taper down in inflation, down to 2%, but if the later explanation is true we are throwing rate hikes and QT at historical events!

The research implies money supply expansion is a relatively small factor - so absent any additional one-time events we can expect inflation to continue plummeting. Perhaps our foot is on the brake a bit hard, considering that supply chains are back to optimized and the stimmie money was spent long ago. This is where "data dependency" could bite the U.S.

[MustacheAndaHalf]

The original "team transitory" was in mid-2021 to late-2021, who claimed that inflation was going to fall back to zero on its own.  That didn't happen, and the Fed admitted "transitory" was wrong in late 2021.  Despite that, people talked of transitory again in early 2022... and inflation spiked higher, rather than going away in 6 months.

If someone guesses inflation will go away in 6 months, they may eventually be correct.  But it helps to look back at "team transitory" track record, which is terrible.

Transitory in 6 months was a failed call, but what if they had said "in 12-24 months we'll see inflation plummeting due to a restoration of supply chains"? Well, that kinda happened too, starting in June 2022. I suspect the timeframe was supplied by the media, rather than the economists. It's also worth noting how some of the economists and influencers who were right in June 2021 about the inflation risk were wrong in June 2022 about the onset of disinflation.

The studies cited in the video are an interesting attempt to tease apart the pandemic supply constraints versus the stimulus-induced boom in demand - both one-time events that self-resolved. The explanation matters because the US had much more stimulus than the EU, and yet the EU had a similar boom in inflation, especially around the Ukraine war. So was it supply? The research suggests different explanations for the two economies' near-simultaneous experiences of inflation.

Most of the factors affecting inflation in the US are in the rear-view mirror (supply chains and stimulus), and resuming student loan payments will resemble an anti-stimulus or even more QT. We really need to know whether inflation is due to money supply remaining far above pre-pandemic levels or if we just went through a stretch of one-time supply and demand events. The Fed seems to be expecting a multi-year taper down in inflation, down to 2%, but if the later explanation is true we are throwing rate hikes and QT at historical events!

The research implies money supply expansion is a relatively small factor - so absent any additional one-time events we can expect inflation to continue plummeting. Perhaps our foot is on the brake a bit hard, considering that supply chains are back to optimized and the stimmie money was spent long ago. This is where "data dependency" could bite the U.S.

I do not view YouTube as a high quality source of information.  The channel you cited, Money & Macro, is run by a PhD student.  He does not list degrees or experience, so its unclear if he has any relevant degrees or experience.  In my experience, non-experts make many mistakes and flawed assumptions, and if you read their material you can pick up their flaws.

I pointed to why "team transitory" lacks credibility.  Are you trying to claim they're credible because they might have said something else?  They didn't.  If you want to use 20/20 hindsight and claim June 2022 was special, in retrospect that was the month of peak inflation.  Predicting what happened over a year ago is much easier than predicting the future, but it does not restore the lack of credibility from team transitory.

If you're a fan of team transitory, you might find Paul Krugman, who won the Nobel Prize in Economics, a higher quality source of information.

[ChpBstrd]

The original "team transitory" was in mid-2021 to late-2021, who claimed that inflation was going to fall back to zero on its own.  That didn't happen, and the Fed admitted "transitory" was wrong in late 2021.  Despite that, people talked of transitory again in early 2022... and inflation spiked higher, rather than going away in 6 months.

If someone guesses inflation will go away in 6 months, they may eventually be correct.  But it helps to look back at "team transitory" track record, which is terrible.

Transitory in 6 months was a failed call, but what if they had said "in 12-24 months we'll see inflation plummeting due to a restoration of supply chains"? Well, that kinda happened too, starting in June 2022. I suspect the timeframe was supplied by the media, rather than the economists. It's also worth noting how some of the economists and influencers who were right in June 2021 about the inflation risk were wrong in June 2022 about the onset of disinflation.

The studies cited in the video are an interesting attempt to tease apart the pandemic supply constraints versus the stimulus-induced boom in demand - both one-time events that self-resolved. The explanation matters because the US had much more stimulus than the EU, and yet the EU had a similar boom in inflation, especially around the Ukraine war. So was it supply? The research suggests different explanations for the two economies' near-simultaneous experiences of inflation.

Most of the factors affecting inflation in the US are in the rear-view mirror (supply chains and stimulus), and resuming student loan payments will resemble an anti-stimulus or even more QT. We really need to know whether inflation is due to money supply remaining far above pre-pandemic levels or if we just went through a stretch of one-time supply and demand events. The Fed seems to be expecting a multi-year taper down in inflation, down to 2%, but if the later explanation is true we are throwing rate hikes and QT at historical events!

The research implies money supply expansion is a relatively small factor - so absent any additional one-time events we can expect inflation to continue plummeting. Perhaps our foot is on the brake a bit hard, considering that supply chains are back to optimized and the stimmie money was spent long ago. This is where "data dependency" could bite the U.S.

I do not view YouTube as a high quality source of information.  The channel you cited, Money & Macro, is run by a PhD student.  He does not list degrees or experience, so its unclear if he has any relevant degrees or experience.  In my experience, non-experts make many mistakes and flawed assumptions, and if you read their material you can pick up their flaws.

I pointed to why "team transitory" lacks credibility.  Are you trying to claim they're credible because they might have said something else?  They didn't.  If you want to use 20/20 hindsight and claim June 2022 was special, in retrospect that was the month of peak inflation.  Predicting what happened over a year ago is much easier than predicting the future, but it does not restore the lack of credibility from team transitory.

If you're a fan of team transitory, you might find Paul Krugman, who won the Nobel Prize in Economics, a higher quality source of information.

I agree about YouTube in general (and Xitter especially), but the proprietor of this particular channel always links to primary sources and generally offers a balanced analysis of the pros and cons of various schools of thought - earning my follow.

These primary research links appear in a blog post which serves as the script for the video. The original research being cited includes authors such as Adam Shapiro, an economist at the Federal Reserve Bank of San Francisco. Cited examples of Shapiro's work include:

https://www.frbsf.org/economic-research/indicators-data/supply-and-demand-driven-pce-inflation/
https://www.frbsf.org/wp-content/uploads/sites/4/wp2020-29.pdf

The general approach of using transactions to determine what percentage of inflation is supply-led or demand-led is repeated by European Central Bank authors such as Pablo Pasimeni, Educardo Goncalves, and Gerritt Koester.

So we're offered a good layman's summary of an emerging trend in the research which underlies central bank thinking. You can evaluate the primary sources, and their sources, for methodological soundness and theoretical coherence. I tend to think these factors matter more than the internet fame or status awards in an academic culture some describe as toxic and non-meritocratic.

We're casting about for ideas to explain inflation's behavior and to forecast central bank moves, so I appreciate being connected to lines of thought and research being conducted by influential economists, even if that bird dogging is done by a graduate student with a finger on the pulse of the profession.

For example, another cited study, published just three weeks ago by the IMF, illustrates the historical realities keeping Jerome Powell up at night:

This paper identifies over 100 inflation shock episodes in 56 countries since the 1970s, including over 60 episodes linked to the 1973–79 oil crises. We document that only in 60 percent of the episodes was inflation brought back down (or “resolved”) within 5 years, and that even in these “successful” cases resolving inflation took, on average, over 3 years. Success rates were lower and resolution times longer for episodes induced by terms-of-trade shocks during the 1973–79 oil crises. Most unresolved episodes involved “premature celebrations”, where inflation declined initially, only to plateau at an elevated level or re-accelerate. Сountries that resolved inflation had tighter monetary policy that was maintained more consistently over time, lower nominal wage growth, and less currency depreciation, compared to unresolved cases. Successful disinflations were associated with short-term output losses, but not with larger output, employment, or real wage losses over a 5-year horizon,

You won't hear about the historical context in a press conference, but this paper explains why the Fed isn't cutting the FFR anytime soon or by any significant amount, now that real rates are in the high end of their range. It is the lesson of the 1970s distilled and empirically catalogued. It reinforces the idea we should be thinking in terms of years instead of months. It suggests even a recession might not persuade the Fed to cut rates back to pre-2022 levels, despite what the buyers of real estate, long-duration bonds, and stocks seem to believe.

[dividendman]

The September jobs report was a blowout at 336k jobs added. It might be upwardly revised just like the August report... are we in for another rate hike or two? Seems like it. Wage gains keep on chugging along.

The 30 Year Treasury is almost at 5% again with the 10 year not far behind at 4.8%, if they get up another half percent or so the yield curve will have un-inverted!

[EscapeVelocity2020]

If the Fed is serious about 2% (which has been a strong and constant message from Jerome), then I'm expecting one more rate hike before the end of this year.  The scary reality about having such a strong labor market (and unions striking, underscoring the upward wage pressure on the supply situation) as well as unrealized asset wealth (e.g. home price inflation) - if the Fed relaxes too soon, there are plenty of underlying forces that would keep inflation higher than 2% for longer.  The time to kill inflation is now, even if a few things start to break.  It would be awfully nice if 'full employment' and job creation would cool off.

[ChpBstrd]

The September jobs report was a blowout at 336k jobs added. It might be upwardly revised just like the August report... are we in for another rate hike or two? Seems like it. Wage gains keep on chugging along.

The 30 Year Treasury is almost at 5% again with the 10 year not far behind at 4.8%, if they get up another half percent or so the yield curve will have un-inverted!

Oof! That is a blowout. Odds of a November 1 rate hike went up from 20% yesterday to 30.6% today. Strangely, stocks are taking the surprise news in stride. This is despite Michelle Bowman, Loretta Mester, and Jerome Powell making recent statements about the likely need for more rate hikes if we see more results like exactly this.

I suspect at least 5/11 votes* are in the bag for a November rate hike** based on the past 3-4 months of strong economic growth numbers, but strangely stock markets seem to be taking the news in stride today. That means the Fed's decision probably hinges on next Thursday's CPI report and the October 27 PCE report.

I always look at commodities indices as a likely indicator of the direction inflation will go. The S&P GSCI was up 2.64% in September, so I'm predicting readings on the high side. If that occurs, it could further seal the deal for a November rate hike.

However, a lot could happen in the next 3 weeks, including bank failures. If anything at all breaks, that rules out a rate hike.

*I predict Kashkari, Bowman, Waller, Logan, and Powell will prefer a rate hike, with the following caveat...

**In practice, I suspect the Federal Reserve likes to show solidarity. If a rate hike clearly isn't going to pass some hawks will change their votes to no and create a false sense of unity rather than raising the questions raised by a 6/5 decision or a decision where the chair is in the minority. Recent lopsided voting patterns suggest they do a straw poll prior to the actual vote and then a couple of voting members realign with the majority.

[ChpBstrd]

Producer Prices went up +0.5% in September, exceeding expectations. On an annualized basis, producer prices increased +2.2%. Excluding food, energy, and trade, PPI rose only +0.2%.

PPI has historically been a leading indicator of the direction of CPI inflation. This plus a rapid increase in the GSCI commodities index suggests tomorrow's CPI number will be on the high end.

[MustacheAndaHalf]

Nowcast CPI estimates 3.4% with core CPI 4.2%
https://www.clevelandfed.org/indicators-and-data/inflation-nowcasting

Those are lower than 3.7% CPI and core CPI 4.3% from last month
https://www.bls.gov/news.release/cpi.nr0.htm

In isolation, I expect the CPI print to be good news for markets (and the "soft landing" narrative).  But I'm not convinced 'in isolation" will be true this week.

[reeshau]

Core Monthly CPI a year ago was 0.573%, so even if it matched PPI, the annual rate will go down.

[ChpBstrd]

Core Monthly CPI a year ago was 0.573%, so even if it matched PPI, the annual rate will go down.

This seems to be the consensus of the FFR futures market. The odds of a November 1 rate hike are down to 8.5%. That means a rate hike is unlikely, but it also means if a rate hike does occur it will shock markets.

Flight to safety from the latest Arab/Israeli conflict slightly pulled down long-term interest rates, which boosted stocks. In hindsight, perhaps uncertainty from the Ukraine war is what drove long-term rates so low for most of 2022 and early 2023.

[ChpBstrd]

Minutes from the Fed's September meeting indicated:

1) Most members think we should get another rate hike this year. Specifically: "A majority of participants judged that one more increase in the target federal funds rate at a future meeting would likely be appropriate, while some judged it likely that no further increases would be warranted."

2) An entire paragraph was devoted to market expectations for future interest rates, suggesting an interest in not surprising markets: "Regarding expectations for the September FOMC meeting, the manager noted that responses to the Open Market Desk's Survey of Primary Dealers and Survey of Market Participants and market pricing all pointed to a virtual certainty of no change in the policy rate. In addition, modal expectations for the policy rate from the surveys were that the current target range would be maintained until the May 2024 FOMC meeting, compared with March in the previous survey, with a roughly one-in-three chance of a 25 basis point increase by the November FOMC meeting. For horizons beyond the middle of next year, the modal path from the surveys was notably lower than the market-implied path."

3) There was more normalization of measuring core inflation without the housing component: "participants also noted that significant progress in reducing inflation had yet to become apparent in the prices of core services excluding housing." Is this an admission that housing is in a bubble?

4) The peak Federal Funds Rate is probably less important than the time until the next rate cuts: "A few participants noted that the pace at which inflation was returning to the Committee's 2 percent goal would influence their views of the sufficiently restrictive level of the policy rate and how long to keep policy restrictive. Several participants commented that, with the policy rate likely at or near its peak, the focus of monetary policy decisions and communications should shift from how high to raise the policy rate to how long to hold the policy rate at restrictive levels."

5) Remember when I said the Fed was likely to leave QT on autopilot after they hit the peak federal funds rate, and maybe even after they start cutting rates? Here's confirmation members of the Federal Reserve are thinking this way: "Several participants noted that the process of balance sheet runoff could continue for some time, even after the Committee begins to reduce the target range for the federal funds rate." If QT is stronger medicine than the Fed thinks, this could result in a policy stance that gets more restrictive, on net, amid slowing economic conditions, despite rate cuts. Such a policy misjudgement could lead to a more severe recession.

[MrGreen]

September CPI Report

CPI-U up 0.4% in September
CPI-U @ 3.7% TTM
Core CPI up 0.3% in September
Core CPI @ 4.1% TTM

[ChpBstrd]

September CPI Report

CPI-U up 0.4% in September
CPI-U @ 3.7% TTM
Core CPI up 0.3% in September
Core CPI @ 4.1% TTM

The consensus forecast for both Core and headline CPI were 0.3%. PPI/commodities again predicted higher-than-estimated core inflation, but the trick didn't work for total CPI inflation this time. Watch for upward revisions though... three one-hundredths of a percent would result in upward rounding for core CPI!

Odds of a November 1 rate hike rose only slightly, but the odds of rates being higher in December went from 28.22% to 38.16%.

Monthly CPI data can be noisy, but we're now into month 3 of an uninterrupted string of rising monthly CPI:
June: +0.15756%
July: +0.15958%
August: +0.2785
Sept: +0.32293

The last such 3-month stretch of increases was December 2022 to February 2023. Prior to that, the last 3 month increasing stretch was early 2021. What's disconcerting about our current stretch is how it's happening despite what most people think is a period of tight economic policy and high interest rates. Inflation growth is not expected to accelerate into such a stiff headwind; it's expected to go backward.

Somehow, shelter costs are still rising. The BLS report says shelter comprised 70% of the increase in Core CPI over the past 12 months. Both rent and owner's equivalent rent rose 0.6% in September, according to the detailed tables. Now consider how shelter components of CPI/PCE lag behind market action by 12-18 months.

This again raises concerns about shelter prices being detached from fundamentals like interest rates and about the possibility that the Fed is attacking a housing bubble with interest rate hikes that affect the entire economy. If that's the case, rates will rise until the bubble is popped, regardless of collateral damage. This all seems eerily similar to the situation in 2007. As was the case then, banks' bond portfolios have already been damaged by rate hikes, leaving them less room to handle additional types of defaults, and the next shoe to drop is an increase in defaults and unemployment as the rate hikes do their work. The main differences today are that the increases in home prices and increases in interest rates are bigger than before the GFC, and that technological changes have left many office spaces vacant even in a full employment economy (so just imagine what a long slump would do).

CPI-U minus shelter hit an annual rate of 2% in September. Mission accomplished except for housing!

If we have a real estate bubble problem and not an inflation problem, we can expect the bubble to burst and rates to fall within the next couple of years. I think the evidence is lining up behind this conclusion. "Higher for longer" could mean the FFR only falls to 2% or 3% amid the next recession.

This point of view is making me take a hard look at going long TLT and ZROZ, or buying very long duration treasuries directly, like the one I'm following in the "most intriguing investment" thread. TLT went from about $85 in mid-2007 to $122 in December 2008.

Now does not seem like the time to reach for yield, given the risks and modest spreads. Instead it seems like the time to reach for duration with top-credit-quality assets and ride convexity back up, or at least not lose money. I cannot recommend buying a house, bank, or REIT amid signs of a real estate bubble that is already in the process of bursting.

[reeshau]

I cannot recommend buying a house, bank, or REIT amid signs of a real estate bubble that is already in the process of bursting.

I view the housing market as more of a "TILT" condition.  So much of the existing market is gummed up by owners locked into their low mortgages that niches are expressing themselves more strongly.  New sales, typically 10% of the market, are 30%.  The people who are left as buyers either are cash buyers, or forced to buy by life circumstances.  In those cases, square footage for new homes has shrunk, making the financing affordable, but lowering the buying price.

I do own a homebuilder, which I began buying in 2020.  One of my major wins this year.  I do sweat that the housing market will get out of this situation in as unpredictable a way as it got in.  I don't think marginally higher rates will change the market much.  What it will take is lower rates to free up existing homebuyers.  Then the question is, how much lower?

All of this is occurring after 11 years of sever underbuilding, plus migration south.  But there is plenty of room for volatility and heartburn in the short run, even in the face of that.

[Financial.Velociraptor]

I did Econ as an undergrad and was at one time a strict Monetarist.  It has become clear since then (1998!) that money and inflation are a lot more squirrely than Friedman ever imagined.  And I have to admit, I just don't "grok" any more. 

I've gone as far as reading Mosler (MMT), which at one time was anathema to me.  What I found was Mosler was data driven rather than theory driven.  He TRIED to apply Friedman school analysis to his work as a sovereign bond trader and it kept 'dorking him up'.  Real markets proved to be lot 'stickier' than the conventional theories held.  And inflation probably required a supply side shock to materialize.  Mosler built new theories based on how markets actually behaved (and made a lot money while Monetarists scratched their heads in denial).  At least, Mosler was vindicated when there was QE to infinity with trivial inflation, up until the point of COVID and China supply shocks! 

This jibes with what is being discussed above about housing.  We built the ever loving eff out of housing building up to GFC (when you could get a new construction loan with no job, income, or assets).  New construction has steeply lagged new household formation for more than a decade (supply shock!) and low and behold, we are seeing sector inflation that monkeying with rates does little to slow.  Per Mosler, it isn't the money supply (or it is but not realized until...Supply imbalance).   

That said, my take is the only thing that will bring headline inflation down is increasing new housing inventory until the vacancy rate returns to the long term average.  Raising rates makes building and buying new homes more expensive and is in fact COUNTERPRODUCTIVE in the current scenario (if you believe Mosler).   Curiously, what the economy needs is not restraint but completely unhinged greed from home-builders. 

[ChpBstrd]

Today I had $50k in US treasury notes mature. To redeploy the cash, I decided to go long… duration that is! I sold 4 put options on TLT for $517 and have a limit order to sell 1 put option on ZROZ for another $48.This is a bold move considering how these have been falling knives for months.

I think we’re probably near the end of the rate hiking cycle and I suspect geopolitical instability will continue to drive demand for treasuries. Chaos in Ukraine and Gaza creates an ideal opening for China to invade Taiwan or for Iran to close the straits of Hormuz to protest Israel’s treatment of Palestinians.

At the same time, we’ll likely have another potential government shutdown in November, which could reduce the supply of treasuries, pushing prices up and yields down. That will have to happen unless Jim Jordan develops a political deathwish.

Finally, banks have suffered fresh damage to their bond portfolios, due to rising long-term rates. I would not be surprised to see a couple more failures in the next six months, as rising delinquencies force some of the smaller banks to recognize losses in HTM portfolios. Stocks like BAC, GS, and HBAN are back to the lows of late April, which I consider to be a reflection of their worsening situation.

Any or none of these outcomes could happen, and a long-duration bond position could still win in the long run. Policy is restrictive and inflation is only being propped up by what I consider to be a fairly obvious housing bubble.

Depending on the squiggles I might buy these ETFs and hold them through the next few months. Dot plot be dammed, the Fed could be cutting rates by the middle of next year if the metrics start looking scary or if externalities occur.

That said, my take is the only thing that will bring headline inflation down is increasing new housing inventory until the vacancy rate returns to the long term average.  Raising rates makes building and buying new homes more expensive and is in fact COUNTERPRODUCTIVE in the current scenario (if you believe Mosler).   Curiously, what the economy needs is not restraint but completely unhinged greed from home-builders.

I suspect Friedman (who has been debunked in a number of ways) will be proven right in his quote about monetary policy working on long and variable legs.

Rates have increased dramatically in the past 19 months, but real rates have only been positive since earlier this year. Meanwhile long-term rates have only in the past few months risen above the current level of inflation. It’s fair to say interest rate policy was still stimulative until a few months ago. I think QT was the only part of Fed policy that was restrictive between mid-2022 and early 2023.

It took over a year for a 0.25% Fed funds rate, most manufacturing facilities being shut down, and trillions of dollars in helicopter money to translate into rising prices for shelter and other components of CPI. Interest rate policy has been above CPI for.. what… 6-8 months now? The lag Friedman was referring to is the difference between what quantitative models say will happen the instant a change occurs and the time it takes consumers, businesses, and governments to actually adjust their behavior.

Prior to the GFC of 2008-09, the Federal Funds Rate hit its peak of 5.25% in August 2006, 16 months until the recession started and a long time before home prices started falling. I think we know where we are in the economic cycle. It’s just very hard to patiently maintain that view as the evidence for economic growth keeps piling up month after month, quarter after quarter. I like to look back and remind myself it has always looked like growth before a recession (kinda the definition).

So in summary, be careful with extrapolating short term trends. I do it too, and have to remind myself to be cautious and humble.We are arguably just now approaching the point where- historically speaking- we might start to see some of the earliest effects from positive real rates.))

[ChpBstrd]

Here’s an interesting speech by Austan Goolsbee, president of the Chicago Fed.

https://www.chicagofed.org/publications/speeches/2023/september-28-peterson-institute

Key points:
-GDP growth, unemployment, and inflation have behaved the opposite of historical norms.
-This calls into question traditional views of a growth/disinflation trade off.
-Goolsbee attributes the inflation surge to supply chain disruptions and the rapid disinflation to confidence in the Fed.
-Stimulus or QE are not mentioned.

[wageslave23]

Here’s an interesting speech by Austan Goolsbee, president of the Chicago Fed.

https://www.chicagofed.org/publications/speeches/2023/september-28-peterson-institute

Key points:
-GDP growth, unemployment, and inflation have behaved the opposite of historical norms.
-This calls into question traditional views of a growth/disinflation trade off.
-Goolsbee attributes the inflation surge to supply chain disruptions and the rapid disinflation to confidence in the Fed.
-Stimulus or QE are not mentioned.

At this point, the fed should just keep their opinions to themselves.  Because whenever I hear them open up their mouth, all I hear is "we clearly have no idea, but we are expected to act like we know what we are doing".

[BicycleB]

Here’s an interesting speech by Austan Goolsbee, president of the Chicago Fed.

https://www.chicagofed.org/publications/speeches/2023/september-28-peterson-institute

Key points:
-GDP growth, unemployment, and inflation have behaved the opposite of historical norms.
-This calls into question traditional views of a growth/disinflation trade off.
-Goolsbee attributes the inflation surge to supply chain disruptions and the rapid disinflation to confidence in the Fed.
-Stimulus or QE are not mentioned.

At this point, the fed should just keep their opinions to themselves.  Because whenever I hear them open up their mouth, all I hear is "we clearly have no idea, but we are expected to act like we know what we are doing".

Or maybe they could continue sharing their thoughts, thus making the Fed one of the most transparent policy setting bodies in the world's halls of power, and contributing significantly to the respect for the US dollar, thereby stabilizing the US and global financial systems.

But hey, why work to build trust while publicly attempting to solve difficult problems, when instead they could just pretend to be all knowing?

The Fed is not supposed to be omnipotent and have all the answers. The best such a body can achieve is to provide reasons for their actions and thereby build greater trust than would likely occur if they were a black box. Also their public deliberations allow thousands of thoughtful investors a better understanding.

At least, that's what I thought was happening, and was part of what this thread is about. Personally I think it's really interesting to see them publicly struggle with cases that challenge their assumptions.

[Radagast]

That said, my take is the only thing that will bring headline inflation down is increasing new housing inventory until the vacancy rate returns to the long term average.  Raising rates makes building and buying new homes more expensive and is in fact COUNTERPRODUCTIVE in the current scenario (if you believe Mosler).   Curiously, what the economy needs is not restraint but completely unhinged greed from home-builders.

That is a good thought, and I will use it as my default assumption going forward. It has the same vibes as a post I read on another forum in 2014 observing that Quantitive Easing could not be inflationary, which from here seems true enough.

I'm not enough of an economist to seek out and read Mossler I think though.

[MustacheAndaHalf]

I've previously heard investors and even professor Jeremy Siegal talk about flaws in the way CPI-U tracks inflation.  But this week I heard a view like that backed by data.

A claim I heard was that 95% of mortgages are fixed rate, which sounds too high to believe... but is confirmed in this article.  Let me round that off, and say all residential mortgages are fixed rate.  Those mortgages are immune to inflation - they're fixed rate.
https://www.financialsamurai.com/adjustable-rate-mortgages-as-a-percentage-of-total-loans/

The Fed tries to represent mortgages using a measure called "owners' equivalent rent of residences", which is 25.6% of the CPI-U.  This measure relies on a fiction: the Fed treats mortgages as owners paying themselves rent.  The CPI-U shows this measure had 7.1% inflation over 12 months, and 0.6% over the most recent month (7.2% annualized).
https://www.bls.gov/news.release/cpi.t01.htm

In other words, one of the largest measures of CPI-U is pure fiction.  Instead of 7.1% inflation, that measure should be 0%.  Fixed rate mortgages have no inflation.  What happens if I try to remove the Fed's fiction?

The CPI-U showed 3.7% inflation over 12 months, with "owners' equivalent rent" having 7.1% inflation over that time.  If I treat that as a math problem, I can remove it from CPI-U inflation:
1/4th of 7.1% + 3/4th of 2.6% = 3.7%

Which means if I set "owners' equivalent rent" to 0% (representing fixed rate mortgages with 0% inflation), I get CPI-U of 2.6%.  If CPI is actually 1.1% lower, that would help explain loose financial conditions and stronger than expected consumer spending.

[BicycleB]

But over time, there is inflation because some of the cheap mortgages do expire, and some new mortgages get made at high new rates. I assume the Fed is trying to approximate this. Is “zero” truly a better model?

[maizefolk]

One can certainly argue that "owner's equivalent rent" isn't useful information and drop it from the index. But instead setting a quarter of the total price index to 0 inflation seems even more flawed than the current method for the following reasons:

A) "Owner's equivalent rent" doesn't only capture principle and interest, but also all the other costs property owners bear (whether landlords or just homeowners). The costs of insurance, property taxes, and maintenance have all increases substantially, whether or not ones principle and interest payments remain fixed.

B) People move. Even in weird economies with high interest rates, people move. Last month 600,000 people give or take bought new houses, likely at new mortgage interest rates which are much higher, and often but not always has higher property values than the last time that bought a house (if they bought a house previously).

If I bought the house I live in now today, which I bought nine years ago, the cost of my mortgage payment would roughly double relative to what I currently pay. If 0.7% of US homeowners see their mortgages double while still living in equivalent houses each month, that would represent 8.7% annual inflation in principle and interest payments.

[FIPurpose]

And homeownership is only about 2/3 of the population, so 1/3 of the population is feeling the full brunt of 7% rent inflation. It creates a dichotomy where people who own are effectively feeling close to no inflation, and those renting or buying today are taking on double that amount.

That may be part of the reason we see such a political divide over inflation. Some people are seeing it, some don't. I'm currently renting out my primary home, but if I compare to today's rates, I'm saving ~16k a year in interest on the home, which basically covers all repairs and then some. It feels crazy to sell it when the differential on the mortgage rates are providing a chunk of free money every year.

And I don't even need to pay down the mortgage with basically any term bond paying more than my rate. I get tax advantages on the mortgage interest, and also get another 1-2% free bonus for "paying down my mortgage" through holding bonds.

[FIPurpose]

Could also take an average rate. If the average home owner only keeps a house about 7 years, you could smooth that number over 7 years, but it'd be an average, not necessarily what people are experiencing, but that's a general problem with CPI across the board.

[Financial.Velociraptor]

Historical Treasury yields:  https://home.treasury.gov/resource-center/data-chart-center/interest-rates/TextView?type=daily_treasury_yield_curve&field_tdr_date_value_month=202310

Labor force stats: https://data.bls.gov/timeseries/LNS11300000

Non-farm payroll: https://data.bls.gov/pdq/SurveyOutputServlet?request_action=wh&graph_name=CE_cesbref1

[MustacheAndaHalf]

One can certainly argue that "owner's equivalent rent" isn't useful information and drop it from the index. But instead setting a quarter of the total price index to 0 inflation seems even more flawed than the current method for the following reasons:

A) "Owner's equivalent rent" doesn't only capture principle and interest, but also all the other costs property owners bear (whether landlords or just homeowners). The costs of insurance, property taxes, and maintenance have all increases substantially, whether or not ones principle and interest payments remain fixed.

B) People move. Even in weird economies with high interest rates, people move. Last month 600,000 people give or take bought new houses, likely at new mortgage interest rates which are much higher, and often but not always has higher property values than the last time that bought a house (if they bought a house previously).

If I bought the house I live in now today, which I bought nine years ago, the cost of my mortgage payment would roughly double relative to what I currently pay. If 0.7% of US homeowners see their mortgages double while still living in equivalent houses each month, that would represent 8.7% annual inflation in principle and interest payments.

The items you mention in (A) are much smaller than mortgage payments.  When combined with a mortgage with 0% inflation, those amounts are going to have net inflation much closer to 0% than 7%.

You claim "have all increases substantially" in a list that includes property taxes.  But property values are only up 1.1% over 12 months according to Zillow (other sources are not up to date - Zillow has Sept 2023 and Sept 2022 data to compare).  Property taxes are based on property values, so I doubt your claim of a substantial increase.
https://fred.stlouisfed.org/series/USAUCSFRCONDOSMSAMID

I agree people are still moving (B), but less often.  Homeowners now need to consider not just a new vs current house, but new mortgage cost vs current fixed rate.  Moving is a factor, but less so because people feel trapped by their much lower mortgages.

[MustacheAndaHalf]

And homeownership is only about 2/3 of the population, so 1/3 of the population is feeling the full brunt of 7% rent inflation. It creates a dichotomy where people who own are effectively feeling close to no inflation, and those renting or buying today are taking on double that amount.

Shelter's 34.749% of the CPI-U is divided into two categories :
"rent of primary residence" : 7.59%
"owners' equivalent rent for residences" : 25.613%
https://www.bls.gov/news.release/cpi.t01.htm

So the data I mention separates renters from homeowners.  The graph below confirms 2/3rds homeownership, yet in CPI-U renters only get 22% weight in shelter instead of 1/3rd.
https://fred.stlouisfed.org/series/RHORUSQ156N

[FIPurpose]

And homeownership is only about 2/3 of the population, so 1/3 of the population is feeling the full brunt of 7% rent inflation. It creates a dichotomy where people who own are effectively feeling close to no inflation, and those renting or buying today are taking on double that amount.

Shelter's 34.749% of the CPI-U is divided into two categories :
"rent of primary residence" : 7.59%
"owners' equivalent rent for residences" : 25.613%
https://www.bls.gov/news.release/cpi.t01.htm

So the data I mention separates renters from homeowners.  The graph below confirms 2/3rds homeownership, yet in CPI-U renters only get 22% weight in shelter instead of 1/3rd.
https://fred.stlouisfed.org/series/RHORUSQ156N

Oh nice. I know it's common to question academics with a bunch of caveats that are usually already covered. It sounds like they do the best they can for making such a broad indicator.

[maizefolk]

The items you mention in (A) are much smaller than mortgage payments.  When combined with a mortgage with 0% inflation, those amounts are going to have net inflation much closer to 0% than 7%.

After about ten years of inflation and property appreciation my monthly mortgage payment is roughly 15% insurance, 30% property tax escrow, 25% principal paydown and 30% interest. Call it at 45/55 split between inflating and non-inflating costs.

But what about maintenance and upkeep? If we apply the 1% rule from rental real estate to estimate my maintenance costs that I pay out of pocket but renters pay for as part of the check that they write to their landlord, that'd bring my cost split to:  ~12% insurance, ~25% taxes, ~20% principal paydown, ~25% interest, and ~17% maintenance. Call it a 55/45 split in favor of more inflating costs vs less fixed costs.

You claim "have all increases substantially" in a list that includes property taxes.  But property values are only up 1.1% over 12 months according to Zillow (other sources are not up to date - Zillow has Sept 2023 and Sept 2022 data to compare).  Property taxes are based on property values, so I doubt your claim of a substantial increase.
https://fred.stlouisfed.org/series/USAUCSFRCONDOSMSAMID

In many parts of the USA, including mine, property taxes are based on the ratio of local government spending to overall assessed value of local property in the taxation zone. The good news is that this means appreciation doesn't necessarily produce an increase in property tax owed if all properties appreciate at the same rate. The bad news is that it also means that when government spending increases faster that real estate appreciates, property taxes go up as a percentage of assessed value.

Personal anecdote: the non-principal+interest portion of my monthly mortgage payment is up 8.4%/year over year so it sure seems like higher inflation has also been hitting the cost of building schools, paying teachers, and maintaining roads. At least where I live.

I agree people are still moving (B), but less often.  Homeowners now need to consider not just a new vs current house, but new mortgage cost vs current fixed rate.  Moving is a factor, but less so because people feel trapped by their much lower mortgages.

That's why I looked at current monthly home sales to get a rough estimate of 0.7% of households moving per month. Since those are the numbers after mortgage interest rates have already gone up it already accounts for fewer people being willing to give up their old low rate mortages.

As I said before, current rates of moving/home purchasing would translate into an overall average increase of 8.7%/year in the amount being paid towards principal and interest on mortgages (0% for a lot of people, but doubling for a modest proportion of people). There is a lot of back of envelope math involved so I'm not confident is my exact 8.7% estimate. However, it is hard to come up with a set of assumptions consistent with the data on home purchases, total US households, and the big recent increases in both property values and interest which wouldn't result in the average principle and interest homeowners are paying increasing somewhere between 5-10% annually.

TL;DR version: While I don't know what the exact right number is, between people moving and costs of homeownership which are subject to inflation even with a fixed rate mortgage, it wouldn't surprise me if the right number was somewhere around the 7%/year the fed is estimating. Or anything in the 5-10%/year range really.

But I'm extremely confident the right inflation number for this category is not 0%.

[ChpBstrd]

No, cutting out owner’s equivalent rent sounds a lot like those arguments from people who say “because I don’t buy X, I am immune to increases in the price of X, and therefore CPI as calculated by the US government should not include X.”

The flip side is the person whose lifestyle makes them hypersensitive to the cost of a particular thing, such as how the people who commute 70 miles each way in an F250 are hypersensitive to the price of gasoline, or the way people who eat a restaurants 3+ times per week are sensitive to that item.

That’s not the point of a metric which is supposed to be nationwide and reflect a weighted basket of stuff people are buying The question being asked is are prices in general rising. The government uses a careful and consistent method to select the basket and the weights just so we don’t have to talk about anecdotes.

The people buying houses at today’s prices might think OER is an understatement.

[Zamboni]

The yield curve is still inverted, but the longer term yields are heading up:
https://www.ustreasuryyieldcurve.com/

US treasuries were the backbone of the suggested investing method when Your Money or Your Life was originally written in 1992. Yields were a few points higher then. Obviously 8-9% guaranteed bond yields for 30 years were hard to pass up at the time, even though putting it in an S&P index fund ultimately worked out better at 9.90% . . . or did it? It depended on the fees on your particular fund, although many have quite low fees of ~0.1%, giving stocks the very slight edge. Over time the yields of new treasury purchases fell as rates were cut . . . but we are headed back in the direction where it makes more financial sense again.

https://www.macrotrends.net/2521/30-year-treasury-bond-rate-yield-chart

I'm sitting on a substantial short position right now, but at some point I'd like to lock in more long position. I hate to be in the "Top Is In" crowd, and I'm still buying stocks in a Roth every month, but more and more data is pointing to negative annual declines becoming the stock market norm for the next decade.

Just curious: at what percentage rate would you personally be willing to buy a 10 or 20 year US treasury?

[maizefolk]

US treasuries were the backbone of the suggested investing method when Your Money or Your Life was originally written in 1992. Yields were a few points higher then. Obviously 8-9% guaranteed bond yields for 30 years were hard to pass up at the time, even though putting it in an S&P index fund ultimately worked out better at 9.90% . . . or did it? It depended on the fees on your particular fund, although many have quite low fees of ~0.1%, giving stocks the very slight edge.

Don't forget the tax advantaged treatment of capital gains and (after 2003) qualified dividends. Assuming a single person living on $40k/year of either bond payments or capital gains/qualified dividends, that works out to about a 5% difference in after tax income. Call it an extra 0.4-0.5% drag on bond yields (will be lower for very low income folks, higher for people receiving more investment income in retirement).

[Zamboni]

^Yes, I wish the tax code wasn't so complicated.

Municipal bond yields are subject to absolutely no federal tax in most cases, while federal bond income is treated as earned income . . . even though I would say that federal bond yields should be the picture in the dictionary next to the definition of "unearned" income. That doesn't eve make sense. But, right now that seems to be the situation. And of course the income tax situation will continue to evolve, with taxes much more likely to increase than decrease in the US, making the Roth so important, as long as they don't change the law and take those tax advantages away.

It's all pretty ridiculous that you have to be an expert in the changing tax codes and various financial instrument fees, etc. to even compare investments.

I suppose it means we have the money to even worry about it, though, which is a good problem to have.

[ChpBstrd]

Just curious: at what percentage rate would you personally be willing to buy a 10 or 20 year US treasury?

I'm interested at today's rates. The 10y real yield is about 2%, and if you measure real yield as the difference with implied breakeven inflation rates it's closer to 2.5%. I'm less interested in yield than I am in convexity. I.e. the potential to make 15-25% in capital gains next year or the year after in response to a couple percent of rate cuts is tantalizing.
 
However, I am worried that I'm anchoring on recent, abnormally low, interest rates, which makes today's somewhat average yields look amazing. We all spent 2009-2021 just wishing there was a way to earn 5% without going way out on a limb, and it's hard to escape that mentality. There's no law of nature saying we can't go to 7% or 9%, but the recent direction of supercore inflation suggests there won't be a need.

I'm also wondering if the yield curves will completely un-invert due to long-term yields increasing, instead of by the usual way when short-term rates are cut going into a recession. I don't have a theory for why the yield curve should un-invert rather than staying deeply inverted until the Fed cuts rates. I guess eventually it makes sense for investors to demand more interest for longer duration/more risk, but why do they decide they need more compensation now?

All I do know is yield curve un-inversion has historically been very close to the start date for recessions, either before or after by a few months. And we're suddenly getting close to uninverting:

[EscapeVelocity2020]

One current thing going on that I've never run across before in the US is how many people who own homes saying that they could never afford to live in their current home if they had to buy it in today's market.  I've never lived in California or New York, which were sort of known for this phenomenon, but it's also happened to Houston!  People with existing mortgages are hanging on for dear life, even though home prices are still high.  The people that wanted to move either have already done so or feel like they've missed the boat and have hunkered down. 

It was an interesting discussion above about how people locked in their current home with a fixed mortgage aren't experiencing 'all of the inflation' (owners equivalent rent), but it still influences their decisions (i.e. not to sell, since they worry about buying or renting...).  And so we have renters that are probably the least able to afford inflation suffering the worst and owners/landlords that are weathering this storm and getting ahead - both feeling less inflation (many of their discretionary purchases and services aren't inflating as fast) and gaining outsized income gains from the worst areas of inflation...

[reeshau]

One current thing going on that I've never run across before in the US is how many people who own homes saying that they could never afford to live in their current home if they had to buy it in today's market.  I've never lived in California or New York, which were sort of known for this phenomenon, but it's also happened to Houston!  People with existing mortgages are hanging on for dear life, even though home prices are still high.  The people that wanted to move either have already done so or feel like they've missed the boat and have hunkered down. 

That's an interesting observation, and not what I would have expected to hear, given the cyclicality of the oil industry.  Our neighborhood was in the $200k's in 2016, the last time the bottom dropped out of oil.  It was back into the $300k's when we bought in the summer of 2020.  And now, it is in the low $500k's.  We definitely would not be looking at this house, and maybe not even this neighborhood, if we had to shop now.

And I say that with no mortgage.  The tax level, while lagging market price swings, has certainly been straining enough.  Almost makes me wish for a (balanced) income tax, which I can manage more directly than neighborhood real estate prices.

[EscapeVelocity2020]

It's reminiscent of when we lived in Norway.  Due to chronically strained supply (codes, labor shortages, cost) people basically never sell their homes.  If they move out, they hang on to the appreciated real estate and rent it out.  The worry is that once you sell, sure you pocket the gains, but you might never be able to hop back on the train...  I'm a huge fan of the Fed being willing to cause some pain in order to prevent inflation from becoming entrenched.  It's not a popular stance, but I'm continually impressed by how steadfast Jerome Powell has been on achieving a 2% inflation target, even just this past week - Speech by Chair Powell on the economic outlook at the Economic Club of New York

[reeshau]

It's reminiscent of when we lived in Norway.  Due to chronically strained supply (codes, labor shortages, cost) people basically never sell their homes.  If they move out, they hang on to the appreciated real estate and rent it out. 

That has been Dublin since 2014, too.  Also complicating the equation are capital gains rates generally higher than income.

[Financial.Velociraptor]

That's an interesting observation, and not what I would have expected to hear, given the cyclicality of the oil industry.

Houston's economy is no longer just oil.  We have the world's largest medical center and that has led to a raging biotech space.  NASA is big and the New Space companies are largely based in Houston.  Much more diversification in the local economy than say the 70s or 80s when Oil dominated everything. 

[reeshau]

That's an interesting observation, and not what I would have expected to hear, given the cyclicality of the oil industry.

Houston's economy is no longer just oil.  We have the world's largest medical center and that has led to a raging biotech space.  NASA is big and the New Space companies are largely based in Houston.  Much more diversification in the local economy than say the 70s or 80s when Oil dominated everything.

I know it isn't what it was.  I also have four people on my block of 13 houses who work in oil & gas.  And it is by no means an outlier in my neighborhood, or suburb.

[ChpBstrd]

The worry is that once you sell, sure you pocket the gains, but you might never be able to hop back on the train... 

I've heard those kind of sentiments too, and it's classic bubble thinking.

[maizefolk]

If you need tulips to live (or at least to live your desired lifestyle), it might make sense to hold on to them even in the midst of a bubble. You lose the potential for a big windfall profit, but the market can and often does stay irrational longer than we expect.