79 comments

[ 3.8 ms ] story [ 114 ms ] thread
How am I supposed to interpret this? That we're currently at the same point we were just before the 2008 GFC?

Because that's how I read it, and I hope it isn't a rhyme.

The interpretation is that people are not saving enough. Which means personal debt is expected to rise, consumption is about to slow down, and corporations will likely see lower revenue growth that they have been used to in the past few years.
I think covid made it weird. People were hyper saving and it caused demand to build up. Then people were “fuck it” after things started to return back to normal. However, people will eventually have less money to spend.
> People were hyper saving

i think instead of thinking of it as saving, it's more "unable to spend". Now that they are able to spend again, they do.

There was never an intentional savings spree imho.

It's not just "unable to spend"; public anxiety tends to cause people to put off spending.
That Americans on average have been horrible savers for a generation or two?
On average Americans have gotten dramatically richer over the past generation or two. The average American is doing far better financially than the average German or Swede.

The median is the relatively ugly picture.

The average American has invested a lot in housing and equities, both of which have soared over the past 20-30 years.

The average wealth per adult in the US is ~$550,000. That's just behind Luxembourg for #2 in the world (led by Switzerland at $685,000). That's a shocking average figure for a nation with 250+ million adults. It's closing in on double that of Sweden, and it's 2/3 higher than the UK or France. It's well over double that of Germany. It's also far higher than very affluent Norway. The US Gini figure is also lower than Sweden now (Sweden has become one of the most unbalanced affluent societies due to their horribly failed immigration integration over the past decade or so).

If you're American, you definitely want to be average (or above), not median.

How about the median wealth per adult? ~$107,000 for the US. That's above Austria, Germany, Sweden, Finland, Japan, Singapore, Spain, Italy. And it's below the Netherlands, France, Canada, Norway, UK, etc.

Germany's median wealth per adult is a mere $66,000. They should be asking some serious questions. Greece is at $53,000 and Portugal is at $70,000 for refence. The UK is up at $151,000, with France and Canada around the $135,000 area.

Average is deceiving. A single billionaire and 999 with zero yields an average of one million each.

With so much wealth at the too, average is going to be pulled up even when there are many with relatively near zero.

It's not deceiving, it's quite clear what it represents. Who isn't aware these days that the US is an unbalanced society in terms of wealth? I think even the typical person commenting on eg Reddit knows that story by now (which is why you'll find that - the gap between median and mean - pointed out in every single relevant thread on sites like that).
Who isn't aware? Wrong question. It's understabding how *greatly* that imbalance skews average. Most people don't know cause from correlation. They certainly don't know one billion is one thousand million. It's not safe to assume that everyome on Reddit or HN are well aware of the subtleties.

Note: The discussion was on savings. Abstracting that from individual wealth adds cognitive load.

> Germany's median wealth per adult is a mere $66,000. They should be asking some serious questions.

Isn't that because most of that "wealth" is from inflated prices of houses? Over 50% of germans rent so over the median point, in US only 34%.

> Germany's median wealth per adult is a mere $66,000. They should be asking some serious questions. Greece is at $53,000 and Portugal is at $70,000 for refence. The UK is up at $151,000, with France and Canada around the $135,000 area.

A lot of this is related to differences in (a) rate of home ownership, and general structure of the housing market, and (b) how pensions work. There's some messiness, but in general these measures of wealth don't account for defined benefit pensions (ie a pension where you get X$, inflation adjusted, per month until you die) very well.

Putting money into a savings account that netted you on average 0.2% yield APR is a great example of a dumb fucking decision.

Thank God Americans aren't dumb enough to let their money rot in an account that gives virtually no benefit.

FDIC insurance certainly has value; sure you're not getting much (or any) returns, but, it's just the extreme end of the risk/reward spectrum.
If we were in an environment where the bank failure rate exceeded 1%, then I would say, sure, thats an OK investment, but that high of a BFR has not been seen. If we were then the investment landscape is much worse than FDIC could insure, rendering itself useless and in all likelihood would deplete itself within a few years (depending on which banks and how much capital gets hit of course).

Then again, Americans have other options to park their assets that are far better places of retaining capital AND growth at the same time, housing being one, stocks being another, bonds being the third worst. So FDIC is about as good as BFR will ever lend you, and currently that means what I noted above - a dumb fucking decision.

Hmm, I’m not sure I agree. Everyone who had their money at SVB got it back thanks to FDIC insurance. Bank failures do happen, in the real world. Just because they’re currently rare doesn’t mean a black swan can’t happen
The Personal Savings Rate on this graph includes money saved via investments. It's not just a Savings Account tracker.

They take what people make, then subtract out money spent on goods, services, and taxes, and interest payments. The money left over is what is considered.

Winter is coming.
in fundamental terms, the savings rate needs to be higher than the depreciation rate or the net value of the economy decreases (hard and fast)

negative savings rates are hard and brutal depression territory.

This chart is hard to interpret, because it's depicting the change in a ratio over time, without necessarily showing the trends of the underlying measurements.

Thankfully, there's another chart that's linked from the first chart, representing personal savings rate:

https://fred.stlouisfed.org/series/PSAVERT

So I guess personal savings rate is falling over time (from ~10% between 1960 and 1990, to ~5% in 2000 and beyond), and gross national income is growing (and maybe even accelerating), which is why the ratio of one to the other has a more pronounced decrease.

Yet I wonder how this data squares with this report from last month:

"The US economy's safety net is bigger than anyone thought"

https://www.businessinsider.com/us-household-savings-estimat...

1. Business Insider is a clickbait publication. You should probably find a better source for any information.

2. Savings rate is a point in time metrics -- d/dt if you wish. A safety net would be the 'area under the curve'.

3. The savings rate is dropping because of increased unemployment and increasing inflation.

I like this headline on the economics page of BI: Taylor Swift and Beyoncé won't prop up the US economy much longer, with Americans' COVID-era savings running dry
Unemployment is not increasing. By any measure it is near all time lows. https://fred.stlouisfed.org/series/U6RATE
Just a nitpick, there are several definitions of unemployment. Technically all are increasing (U6, which you linked is +0.3 in the last year) but you're still correct about them being near all time lows. Definitely within normal variance range and I'd be pressed to call it "increasing" from a point of worrying rather than a "well technically" point. (this is why I hate when people are like "x has risen <huge number>%!" -- or decreased -- because it doesn't matter if the numbers are still low. +7.5% increase in unemployment sounds like a lot but it isn't because the baseline is still low. +7% of 6.7% is very different than +7% of 10%)

https://fred.stlouisfed.org/release/tables?rid=50&eid=4773#s...

So is the labor participation rate over the same time period: https://fred.stlouisfed.org/graph/?g=1b4eJ

As boomers exit the labor market, jobs free up. That may not be the only explanation but it's likely part of it.

> So is the labor participation rate over the same time period

Yes, because the denominator for labor force is the population over 16, on the premise that people age into "working age", but never age out of it, a demographic bulge entering the age range at which people both retire if they can, and frequently become unable to work even if they can't afford to retire, drops labor force participation rate when you don't have a similar demographic bulge entering working age.

The prime age labor force participation rate (the LFPR with a denominator based on prime working age having not only an entrance age of 16 but also an exit age of 55) is not low, in fact, its the highest its been any time since the dotcom bubble burst in 2001. So, yes, the low overall LFPR is entirely explained by "we have lots of old people".

https://fred.stlouisfed.org/series/LNS11300060

Here is a different calculation:

Number of people employed 2019(pre-pandemic)/2023: increase of 1.9%

Population of the USA 2019/2023 (estimates): increase of 3.5%

The situation could be the same of better (ie: it's possible the population is younger and we should account for that); but clearly the job market is doing worse (regardless of population) per-capita than in 2019. This will have inflationary effects (things are more expensive, more disruption of services, etc...) as there are more new people in the country than people working.

The way the U6 rate has been gamified, it'll probably be useless as a metric by the end of the decade.

> This will have inflationary effects (things are more expensive, more disruption of services, etc...) as there are more new people in the country than people working.

It's more or less the opposite (though with some of the same concerns); this is driven by an _aging_ population. The population is going up due to birth and immigration, but faster than that is happening, people are aging out of the workforce.

This is pretty much what all developed countries are facing; the US isn't as far along as most (it's relatively young).

If the GDP grows but the people saving money aren't getting the same cut, then this chart will trend down. The implication being that the middle class is getting to share in the GDP growth as the rich do. However I suppose this could also represent the rich saving fewer dollars and instead buying assets. Would be interesting to see this chart but separated by income class.
Actually median (and poorer) households did well between 2019 - 2022. See https://www.federalreserve.gov/publications/files/scf23.pdf or the related blog post https://www.noahpinion.blog/p/great-news-about-american-weal... . This chart doesn't seem to be about households alone.
Those at the bottom did well due to government "one time" contributions, not because the economic system miraculously raised them.
Yep, probably true, but the government redistributing income is in itself kind of part of our economic system.
> If the GDP grows but the people saving money aren't getting the same cut

This is much more meme than truth, and it's frustrating how many people just take this kind of statement as a prior. Here's a FRED chart of median personal income as a fraction of per capita GDP: https://fred.stlouisfed.org/graph/?g=1b46r

Indeed it has been trending downward, but... not by a lot. In the 80's it was 61% or so (so the median person got 61% of their "fair share" if the GDP was shared equally). Right now it's about 53%, up from a minimum in 2013 of about 52%.

And of course in exchange real per capita income has grown wildly (up more than 50%) in that same 40 years. If you told someone in 1980 that you could get a 50% raise, but only if the people above you on the ladder were allowed to get a 65% raise, would you have rejected it?

“Not by a lot”? The very chart you linked shows the median “share” dropping like 25% over the last 50 years.
Most people wouldn't look at 25% change in distribution over a half century as a big effect, no. Look at all the other stuff (healthcare or TV prices, say) that have swung by two orders of magnitude. And again, wealth in absolute terms has gone up by twice that!

So, no, not by a lot. I think I stand by that. If you're genuinely upset over 25% fairness in a context where everything else looks so much better, my contention is that you're just looking for something to complain about.

I’m “just looking for something to complain about” because you’re so busy looking for boots to lick
The personal savings rate has a big spike over the pandemic, and household net wealth increased a lot between 2019 - 2022 (https://www.federalreserve.gov/publications/files/scf23.pdf)* which is maybe what that safety net headline is about.

The story seems to be large government deficits balancing out the high savings of households/businesses, which is why the overall savings rate is near zero.

* Median household net worth increased by 35% from 2019 to 2022, adjusted for inflation. For the bottom quartile of households, the median household net worth increased by a whopping 900%.

> Yet I wonder how this data squares with this report from last month: > > "The US economy's safety net is bigger than anyone thought"

Don't those express the same underlying truth? In an economic situation without safety net backstops, citizens have to save in order to survive contractions. In modern america, they don't need to as much. So they don't.

FWIW this is a real problem with "savings" as a metric. It's defined based on "disposable income", therefore fails to track things that probably "should" be savings but are accounted for separatly. Your 401k match or HSA deduction is not "savings", for example, even though clearly they are, and clearly they represent the kinds of savings that would have shown up in a bank account (and thus have been technically "savings") in the 60's.

Do you know of any data source that stratisfies this by age group over a decently long period of time (say last 5 decades)? I would expect this number to take a nose dive as the ratio of retirees to works shifts towards retirees so i am not sure what conclusions i can draw from the data without some correction for retiree ratio.
Actually what amount to "net saving" in the graph? Is it the amount of money people put in their checking/saving account? The amount including the money put in the investment account? Including retirement?
There is some description and link on the page.

  Personal saving as a percentage of disposable personal income (DPI), frequently referred to as "the personal saving rate," is calculated as the ratio of personal saving to DPI.
  Personal saving is equal to personal income less personal outlays and personal taxes; it may generally be viewed as the portion of personal income that is used either to provide funds to capital markets or to invest in real assets such as residences.(https://www.bea.gov/national/pdf/all-chapters.pdf)
https://fred.stlouisfed.org/release/tables?rid=54&eid=155443...
We're still near the 2000 amount - about 25% below where we were during the GFC;

https://fred.stlouisfed.org/series/HDTGPDUSQ163N

How do I interpret that Q4 to Q1 sawtooth? End-of-year bonuses paying off debt?
Nope - or maybe I guess - but as an input into GDP instead of debt. Non-seasonally adjusted GDP has that same pattern, Q4 is almost always higher than Q1 so the version you usually see for GDP is seasonally adjusted (Red is adjusted, blue is the 'raw' GDP figure).

https://fred.stlouisfed.org/graph/?g=1b3ZV

Q4 always has the highest GDP. presumably because of holiday spending.
Debt (measured in dollars) and GDP (measured in dollars per year) is a strange comparison.
I wonder if 'net saving' here includes businesses/government or just households. Noah Smith wrote a nice piece here https://www.noahpinion.blog/p/great-news-about-american-weal... breaking down Federal Reserve data that shows how net wealth of American households increased a lot from 2019 to 2022 - after adjusting for inflation, and even more so for poorer households.

https://fred.stlouisfed.org/series/PSAVERT - this data series is labelled "Personal Savings Rate" and shows a big spike over the pandemic years, which would square with the increase in household net wealth. Maybe 'personal savings rate' refers to households, and so the story here is a transfer of wealth from the government to households over the pandemic years?

Of course (now I'm really extrapolating) the govt is just funded by taxes, so maybe that further reduces to a transfer of wealth from future households/businesses to current households.

You're correct - it includes all saving -- households, businesses and government - net private saving which is households + businesses has been strong:

https://fred.stlouisfed.org/series/W202RC1Q027SBEA

But Government saving is obviously in a pretty big deficit: https://fred.stlouisfed.org/series/TGDEF

Thanks, that makes sense. I wonder to what extent the two rates are directly linked - household savings coming from direct govt transfers like covid stimulus checks and student debt forgiveness.
This seems to be a good way to visualize inflation, but that's all.
Is gross national income simply all household income?
https://en.m.wikipedia.org/wiki/Gross_national_income

IIUC, GDP counts income owned by foreigners that happened in the US. GNP counts income only owned by residents of the US that happened in the US.

If someone from China owns a toaster making factory in the US, which employs no one, and it generates $100M of toasters a year - that's in GDP but not GNP.

That toaster factory is in US GDP and in China’s GNP. (I think.)
> GNP counts income only owned by residents of the US that happened in the US.

Also income of US residents that happened outside the US, no?

IIUC, "domestic" is location-based while "national" is residency-based.

IIUC, it is income of any "person" of the nation, including personal, business and government.
Everyone was able to save until the massive money printing event the last few years. Same amount of stuff, but a lot more money created ($18T -> $23T) means everything now costs more.

So much more that many things (cars, houses, etc..) in some areas doubled in price and people have spent all their savings to live the same life they had in 2019.

https://fred.stlouisfed.org/series/QBPBSTAS

Median household net worth increased by 35% from 2019 to 2022, adjusted for inflation. For the bottom quartile of households, the median household net worth increased by a whopping 900%.

Pessimism seems to be in vogue right now (or maybe always?) but in this case the data doesn't seem to bear it out.

Define net worth; the vast majority of people live paycheck to paycheck, can't cover a $1000 emergency, and don't actually own the assets they utilize.
Net wealth is basically the value of all the assets you own, minus all the debt you have. The data ( https://www.federalreserve.gov/publications/files/scf23.pdf ) just doesn't support what you said. Even within the bottom 25% of households ('bottom' with reference to their net wealth), the median net wealth increased from $400 to $3500, and this is adjusted for inflation (all in terms of 2022 dollars). Source is p18 of that PDF I linked above
Cite a source that doesn't have a conflict of interest. Inflated home prices and surging used vehicle markets makes for positive data, but not reality.

https://www.wsj.com/articles/pandemic-savings-in-u-s-running...

I agree that a lot of the net wealth increase came from asset prices going up, but your original comment said "don't actually own the assets they utilize" which is not the case as evidenced by the data I linked.

You can also see that debt has fallen, p31 of the report I linked. So that's another mechanism through which median net wealth has increased, which is not asset price inflation.

As far as sources, I trust a technical Fed report more than an opinion analyst piece written by 'economists' in a private bank. I may be biased as a formal central bank economist :) That report is half speculation about the future ("looks set to end soon, with excess household savings likely to be depleted by year-end"), whereas what I linked is actual data from the past.

>I trust a technical Fed report >As a formal central bank economist

So you trust in what you already have faith in? That's not a source that inspires confidence in our discourse.

>That report is half speculation about the future ("looks set to end soon, >with excess household savings likely to be depleted by year-end")

As other pointed out in this thread this data suggests differently. https://fred.stlouisfed.org/series/PSAVERT All trends are speculations about the future. From your own report: "By this metric, in the 2022 survey, 82 percent of families in the top decile of the usual income distribution saved, compared with 66 percent of families in the upper-middle segment and 43 percent of families in the bottom half. Between 2019 and 2022, the overall fraction of families that saved edged down from 59 percent to 56 percent, with decreased saving observed among all three segments."

So savings spiked then disappeared at the same time all these positive numbers on assets began to surface. How interesting.

Yet, "Among families in the bottom quartile, median net worth was $400 in 2019 and $3,500 in 2022, and *mean net worth was negative $15,700 in 2019 and negative $5,300 in 2022.* Among families in the top decile, median net worth was $3,012,500 in 2019 and $3,794,600 in 2022, and mean net worth was $6,641,800 in 2019 and $7,810,500 in 2022."

In addition to, "Younger people – those below the age of 35 – are far more likely to rent than are other age groups: About two-thirds (65.9%) of this age group lives in rentals. This compares with, for example, 42% of those ages 35 to 44, and less than a third (31.5%) of 45- to 54-year-olds." (https://www.pewresearch.org/short-reads/2021/08/02/as-nation...)

So the growth in assets from homeownership doesn't make sense for a large majority of the population; but, for those who rent the increase in personal savings makes a ton of sense given the CDCs moratorium. The increase in upper class wealth makes sense as many bought additional homes; which fits with the ever increasing prices in homes. The decrease in savings and the rising cost of borrowing would indicate trouble ahead. The confidence gained from well manufactured data is fleeting.

> So you trust in what you already have faith in? That's not a source that inspires confidence in our discourse.

I trust central bankers to not straight up make up or lie about facts, because I worked in one and formed an impression of what the institutional culture is like, yes. I did volunteer that information to you though.

The rest of what you're saying sounds like older people have seen increases in net worth because of asset price inflation, and younger people have seen increases in net worth because of covid-related measures ("CDCs moratorium"). That sounds right to me too!

> Everyone was able to save until the massive money printing event the last few years

This isn’t really true [1]. American savings rates have been secularly falling since the 60s, but they’re relatively healthy right now.

[1] https://fred.stlouisfed.org/series/PSAVERT

Are we looking at the same chart? The only other time we've been this low was 2001 after the dot-com and again in 2005-2007 when everyone emptied their bank accounts trying to buy a house.
I think he meant employed people during covid. I was able to save a ton and pay off all debt, just because there wasn't really anything to buy. Now I feel the inflation squeeze.
It'd be interesting to see this compared against the rise of computerized banking and access to credit. Ubiquitous credit card payment didn't really come in until the 1990s
Just to set context here.

Net saving is:

> equal to the sum of personal saving, undistributed corporate profits with inventory valuation and capital consumption adjustments, and net government saving. [0]

To break that down further:

Personal Saving is 'Post tax income less consumption'. That looks like this: https://fred.stlouisfed.org/series/PSAVE

Undistributed corporate profits with inventory valuation and capital consumption adjustments is a measure of how much profit companies made, but did not return to shareholders (with an adjustment to remove the effect of inflation). That looks like this: https://fred.stlouisfed.org/series/B057RC1Q027SBEA

Net Government saving is (oversimplified, but basically) the deficit: https://fred.stlouisfed.org/series/TGDEF

GNI is sort of like GDP, but slightly different (treat it like the same for nearly all intents and purposes).

What this chart is basically showing, is that Federal deficits are up, but it's not translating to a higher household savings rate (As was the case during the stimulus period) OR higher retained corporate earnings.

I'd argue that the actual number is an intellectual exercise (i.e. it's relatively meaningless that it is now negative), but it is worth investigating why the Government is running such a large deficit in a hot economy that is not leading to better quality of life (Personal savings is a rough proxy for that).

[0]https://www.bea.gov/help/glossary/net-saving#:~:text=A%20mea....

So 'post tax income less consumption' is the money after the pretax retirement money put away.
I think it’s really hard to interpret this chart without the wealth element - if I own a $300k home and and I make $50k a year, and I normally save 10% ($5k), and my house goes up in value 10% ($30k), one might feel better about spending (not saving) the $5k as I would still be $30k “richer” (as opposed to $35k).
There are so many moving parts, indices and measurements in economics, anybody sufficiently familiar with the data landscape can "prove" any (typically political) point they want by selectively focusing on some timeseries or another.

This is not to say there is no objective economic reality, but meaningful comparisons over time and across different economies seem very hard. Thats why "economically you never had it so good" can coexist with outer despair.

And this before we account for the non-accounting of externalities and the variable long term sustainability of different lifestyles (the wealth transfer from future generations).

No doubt these all are difficult to interpret. It's still a great thing to have numbers (even when the methodology for some numbers is very questionable).

But if you are trying to drive or merely gauge your investment performance, or your voting strategy, this is hard to read. For example, it's fair that people were able to spend much less during Covid. Bad for vendors, great for people's overall savings which had been often too low for a while. And so, fair that there was a rebound in spending. Which is a problem when it leads to high interest rates and a major step in inflation (which some people deny exists altogether). Meanwhile there was giant increases in money supply. Most of which is not going to be re-absorbed anytime soon if ever. And is bound to have a commensurable effect. Etc, etc, etc.

And so in the end how do I interpret my current investment performance? (And equally interesting question is how should our country spend / invest / repay / drive this or that). Just pick a question, any question. Is most of my investment performance purely driven by increased money supply? (meaning that even if I over-performed indexes, I may have grossly underperformed inflation - i.e. be much poorer). Hard to tell actually. Of course one way is simply to spend conservatively (i.e. too little) and invest "more than needed". But obviously that's grossly sub-optimal. For example if your objective is to spend more than comfortably so as to die pennyless <- it's an exageration but not a bad strategy depending on family or donation objectives. Anyway. Tough job.