Yes There Has Been Inflation

2023-07-03: When I wrote this post, inflation in the US had just started rising and everyone was saying that it was “transitory” and pointed to the past decade of low inflation plus stimulus to justify things would be ok. Since then, there has obviously been a large increase in core inflation in 2022 and folks stopped arguing it was transitory. I still think the points in this post are relevant and worth considering.

I recently read The Deficit Myth and I particularly enjoyed the insight that inflation comes from consumer spending on supply-limited goods, not printing money. Put differently, when a consumer wants a scarce good and has more money to spend they will accept a higher price for the same good. Inflation only occurs when the consumer accepts that higher price.

This realization has some interesting consequences for U.S. Federal monetary and fiscal policy:

Printing money does not inherently cause inflation. People spending money causes inflation and it matters more who has excess money (or credit) to spend.
Spending money to create supply only causes inflation on the materials that supplier requires. It is less likely consumers would see these price increases long term as the increased supply compensates.
For the federal government, tax cuts are functionally equivalent to printing money and handing it directly to certain people. Taxes remove money from the system, spending creates it.

Under this insight the past decade of loose monetary policy but low inflation in the consumer price index (CPI) makes sense if that money supply did not primarily flow to most consumers who buy most goods.

I contend there actually has been inflation, but in goods that wealthy Americans compete for, not for everyday goods. Relatedly, I personally believe this is a sign we as a society are not efficiently distributing wealth to most people, but concentrating it in the hands of the few who then use it to compete for luxury goods.

There has actually been inflation

Most of the money that has been created in the past twenty years via tax cuts, stock market gains, defense spending, and government guaranteed loans (in particular mortgage guarantees and education loans) has flowed to generally wealthier people and not to most consumers.

Wealthy people don’t compete for gallons of milk, they compete for scarce assets and luxury real world goods. For example, a wealthy person might compete with other wealthy people for:

(Fine) Art
(Desirable) Land
(Large) Yachts
(Fancy) Jewelry
(New) Cars
(Higher) Education

Average prices on all of these have increased, but just like measuring average latency doesn’t reveal much information, we want to look at the statistical tail of the distribution to see where the inflation lives.

Unfortunately, there isn’t very much in the way of price distribution data available for these asset classes (say relative to liquid assets like stocks or bonds) and I’m not sufficiently invested in this post to try to get it, so let’s just look as some of the tail events that have happened in the last decade and see what we see. We can calculate inflation rates with a simple function:

def inflation(p, y):
    """
    Given two prices and two years, calculate the annual
    inflation rate as a percentage (0-100)
    """
    p1, p2 = p
    y1, y2 = y
    ratio = p2 / p1
    years = 1 / (y2 - y1)
    return round((ratio ** years - 1) * 100, 2)

# inflation((127.5, 450.3), (2013, 2017)) -> 37.09

So if we were to look at CPI we’d see inflation of around 3% annually since 1980 and even lower around 2.3% since 2000.

# From 1980 to 2021
inflation((77.8, 261.582), (1980, 2021)) -> 3.0
# Or just from 2000 to 2021
inflation((168.8, 273), (2000, 2021)) -> 2.32

In the same period median full-time annual wages have increased as well, around 3.4% since 1980 and 2.7% since 2000. Even the first decile of wages have increased similarly (data is only available since 2000 for that series). Median household income has also increased similarly in current dollars

# Median full-time wages from 1980 to 2021
inflation((254, 990), (1980, 2021)) -> 3.37
# Median full-time wages from 2000 to 2021
inflation((568, 990), (2000, 2021)) -> 2.68
# First decile full-time from 2000 to 2021
inflation((281, 502), (2000, 2021)) -> 2.8

# Median household income from 1984 to 2020
inflation((22_415, 67_521), (1984, 2020)) -> 3.11
# Median household income from 2000 to 2020
inflation((41_990, 67_521), (2000, 2020)) -> 2.4

Let’s focus in on the last two decades, where CPI has inflated by around 2.3% and median household income have increased by around 2.4%. How do these average rates compare to prices at the tail of the distribution?

Desirable Art

The 2010s have been good for fine art, seeing some record breaking sales such as:

2017: $450.3 million for Salvator Mundi by Leonardo da Vinci. Valued at $127.5 million in 2013 for a 37% annual inflation rate.

inflation((127.5, 450.3), (2013, 2017)) -> 37.09

2017: $110.5 million for Untitled by Jean-Michel Basquiat, the largest sum ever paid for an American artist at auction. Valued at $20,900 in 1984 this is a 30% annual inflation rate.

inflation((20_900, 110.5e6), (1984, 2017)) -> 29.67

2015: $179.4 million for Les Femmes d’Alger, Version O by Pablo Picasso. Sold for $31.9 million in 1997 this is a 10% annual inflation rate.

inflation((31.9, 179.4), (1997, 2015)) -> 10.07

As many (most?) sales of art are private it is hard to know exactly how much inflation has been going on in the fine art community. Above we looked at some of the tail events happening in public auctions over the past decade and hopefully as the LLCs created by companies like masterworks.io become more common we can finally get some accurate market data.

That being said, fine art is inflating at a rate far greater than 2.3%, probably closer to 10%.

Desirable Land

Ideally we could break down land value separately from homes so we can look at how the actual scarce good (land) performs relative to the durable good of the house itself. I can’t seem to find good data on per locality land values without homes on them, but Zillow Research does provide the Zillow Home Value Index (ZHVI) for “Top Tier” markets (65% - 95% for a region) which can give us some insight into how the most expensive real-estate markets in America have inflated since 2000.

Homes in Atherton, California (rich Tech Bay Area, part of “Silicon Valley”) have increased from $4.6 million in 2000 right at the peak of the tech bubble to $12.4 million in 2021, for a 4.85% inflation rate. If you just look from 2012 to 2021 the rate increases to around 7%.

inflation((4_599_043, 12_445_835), (2000, 2021)) -> 4.85
inflation((6_804_630, 12_445_835), (2012, 2021)) -> 6.94

Homes in Aspen, Colorado have increased from $3.1 million in 2000 to $8.36 million in 2021, for a 4.8% inflation rate.

inflation((3_125_662, 8_362_437), (2000, 2021)) -> 4.8

Homes in Bridgehampton, New York (part of the “Hamptons”) have increased from $1.82m in 2000 to $7.8 million in 2021, for a 7.11% inflation rate.

inflation((1_842_709, 7_793_749), (2000, 2021)) ->: 7.11

Near where I grew up the most desirable homes were in Potomac, Maryland (rich public figures and such from DC). From 2000 to 2006 homes there increased from $782,806 to around $1.7 million, for a 14% inflation rate, but never really recovered after the financial crisis and stand at about the same value today yielding a 3.6% annual inflation rate from 2000 to 2021.

inflation((782_806, 1_713_201), (2000, 2006)) -> 13.94
inflation((782_806, 1_652_206), (2000, 2021)) -> 3.62

At the top end of the real-estate market we have also seen higher price increases than 2.3%, although it is very time dependent, probably around 5% though.

Desirable Watches

I wasn’t sure if I should include watches in this analysis because desirable luxury watches have had a somewhat large rise in prices since 2019, but I do think it’s a good example of the kind of supply limited luxury goods that wealthy people are absolutely competing for.

For example, the Patek Philippe Nautilus 5711 (blue dial 1A-010) has increased from around $24,215 in 2013 to nearly $145,000 in 2021 according to Chrono24. Yes there are stainless steel watches going for over one hundred thousand dollars on the secondary market! Rare and precious metal watches have increased perhaps even more such as the Rolex Rainbow Daytona ref 116595RBOW which in rose gold were originally sold at $96,900 in 2018 and in 2021 can go for as high as $500,000. Watches that are intentionally kept scarce such as those produced by Richard Mille routinely sell for between a quarter and half a million dollars, often far above retail prices.

inflation((24_215, 145_000), (2013, 2021)) -> 25.07
inflation((96_900, 500_000), (2018, 2021)) -> 72.8

At the top end of the luxury watch market we have seen price increases similar to that of fine art, probably because luxury watches are more fine art than functional. Although data is scarce, it is likely that luxury watches in general have experienced around 10% inflation over the past two decades.

Desirable Education

We often hear about how the cost of higher education is increasing in America for a multitude of reasons. Under the “available money spent on scarce goods creates inflation” philosophy one might explain it primarily because more consumers have access to credit (often with rather punishing terms) to pay for college (demand) while the number of slots at universities (supply) have remained somewhat constant. But at the tail, how have tuitions changed at the top three Universities in the world (according to US News and World Report)?

One year of education at Harvard University has increased from $32,164 in 2000 to $63,025 in 2017 for a 4% inflation rate.

inflation((32_164, 63_025), (2000, 2017)) -> 4.04

One year of education at MIT, which apparently includes much more expensive housing than Harvard, has increased from $33,225 in 2000 to $77,020 in 2021 for a 4.08% inflation rate.

inflation((33_225, 77_020), (2000, 2021)) -> 4.08

One year of education at Stanford has increased from $32,471 in 2000 to $73,333 in 2021 for a 3.96% inflation rate.

inflation((32_471, 73_333), (2000, 2021)) -> 3.96

Even if we look at a large in-state university like Ohio State, we see one year of in-state education has increased from $12,483 in 2000 to $25,288 in 2021 for a 3.4% inflation rate.

inflation((12_483, 25_288), (2000, 2021)) -> 3.42

At the top end of the education market we have again seen higher price increases than 2.3%, probably around 4%. This was somewhat surprisingly low to me, in my opinion, as a 20 year US Treasury Bond yielded around 6% in 2000 (although only 1.5% today), and median household wage growth was around 2.4% over roughly the same period. Bear in mind though that these are annually compounding rates, so over many years small percentage differences can manifest in large outcomes. In addition, there is likely further analysis to do here around the real cost of University as a larger share of college is paid for with debt rather than pure income or savings.

Conclusion

From the price increases of goods that wealthier people compete for, we see about double (in some places more than double) the officially reported rate of inflation of the broadly applicable CPI. This data matches with the hypothesis put forward by the Deficit Myth, namely that spending causes inflation, not money supply.

I by no means think we should sympathize with the inflationary pains of the elite, but I do fear it signals broader political ramifications. In particular, as the price of admission to the upper class of society increases, more potential elites are shut out of upward mobility. I fear we are already seeing this as college degrees devalue, owning homes seem unattainable to many, and financial independence may be out of reach.