House Prices and impossible things
House Prices
There is nothing quite like ‘good ole bricks n mortar’ to warm the cockles of the hearts of the investing public, especially in the UK. Emotional attachments aside, much of this fuzzy warm feeling comes from the fact house prices have gone up substantially in recent decades.
Over the period, inflation has compounded at 3.6% while house prices have increased at a compound 5.7% per annum with the average house price in the UK now around £295,000. This period reflects the experience of virtually all UK homeowners.
At first blush, this all seems pretty reasonable. After all, 5.7% is what we have come to expect and growing by a couple of percentage points above inflation is what we should expect from this most gilded of asset classes. However, it is my contention that this level of price appreciation above the trend in the general price level over the long-run, is not “pretty reasonable”, it is in fact, an impossible thing.
To illustrate, let’s consider the issue from both mathematics and history perspectives. Imagine these returns not over a few decades but over a much longer period, say 500 years. Let’s assume that inflation had been 3.6% for the last 500 years and the general price level had reached where it currently is today. Also, imagine that over this period, UK house prices had also appreciated at 3.6% annually and stood at an average of £295k today. This means that there has been no increase in house prices in real terms in our 500-year period. Now assume that instead of 3.6%, house prices had matched their recent run and appreciated at 5.7% or 2.1% above inflation for 500 years. In this event, the average UK house price today would be not £295k but £6.7 billion in current pounds. Surely an impossible thing.
The power of compounding over long periods (Einstein’s the eighth wonder of the world) mean that NO REAL ASSET can consistently appreciate above the general price level for very long periods otherwise its size, relative to everything else, becomes impossibly large.
Fortunately, we don’t have to rely solely on the maths. Long-term historical house price data focusing on the Herengracht canal area in Amsterdam exists, going back to 1620, along with general prices that allow a real price index to be calculated.
Chart: Valuabl
Created with Data Wrapper
The data shows clearly that real house prices have ebbed and flowed over time, but until recently, largely remained anchored to the general price level. Similarly, rents have moved in lockstep with prices. I will address the explosion in real prices that occurred from the early 1990s onward, later in this article. At this juncture, I will make two observations. Firstly, real house prices presented in this data were the same in 1990 as they were in 1620. Secondly, within the ebb and flow in this long-term data, real prices fell by around 60-70% twice, and sometimes fell for extended periods.
Looking at other (albeit shorter) data sets, Nobel laureate Robert Shiller provides inflation-adjusted home prices in the United States over the past 130 years.
Once again, until the last couple of decades, house prices have broadly moved in line with goods price inflation with real prices in 2000 no higher than 100 years prior.
Buffett’s Folly
A house is a real asset that should be expected to maintain its real value over time. However, because it is not a productive asset, like an equity which combines inputs to generate more valuable outputs, its price cannot deviate materially from the general price level indefinitely. A house is not a productive asset in that its output, a place to live, cannot be stored and is fully consumed every year it is produced. In this way it differs from productive capital, such as plant machinery and intellectual property, created from combining savings with endeavour. Once a house has been built, any value that it accrues (excluding improvements) comes exogenously from an increase in the value of the land on which it sits.
To illustrate the difference, in February 1958 Warren Buffett bought a house in Omaha for $31,500 that he still lives in today. Buffett only bought the house under duress from his wife and made the following quote at the time “In Omaha, I rented a house at 5202 Underwood for $175 a month. I told my wife “I’d be glad to buy a house, but that’s like a carpenter selling his toolkit.” I didn’t want to use up my capital.” Buffett has made significant improvements to the property over the ensuing years, and it is currently worth around $1.4m. Excluding these, the original property would be worth around $650k today. If instead, he had convinced Mrs Buffett to couch-surf at Charlie Munger’s and he had invested the money in Berkshire Hathaway, it would be worth around $6.6bn. Even a boring old S&P 500 tracker would have been worth $18m. Shortly after the purchase, Buffett christened his new abode “Buffett’s folly”.
This analysis is focused on the price level, not the total return if viewed as an investment. Houses that are owned as investments provide a rental yield which is then reduced by financing costs, maintenance costs and tax. These factors can vary over time. For leveraged investors, current rental yields are broadly speaking, below buy-to-let mortgage costs which makes for a negative carry investment and the return a function of the (real) price appreciation (or depreciation). For unleveraged investors, the net rental yield improves the return profile of the investment in isolation but must be considered in terms of the opportunity cost relative to other investment opportunities.
None of this detracts from the fact that housing is a real asset that should not appreciate (or depreciate), in real terms, over long periods of time. All in all, not a terrible deal if you can take a very long-term view and in many ways an excellent enforced savings product. But not at any starting price. When real house prices have been inflated by conditions (the credit cycle) they are highly exposed to a reversion to the mean. Which brings us to today.
The Rise and Rise of the Global Balance Sheet
The November 2021 McKinsey Global Institute report titled “The rise and rise of the global balance sheet.” goes into detail on the increase in wealth in absolute terms and relative to GDP that has occurred across the globe since 2000. The global value of net wealth has more than tripled since 2000 from $150tn or 4 times GDP to about $500tn or 6 times GDP in 2020. This statistic alone sounds a lot like an impossible thing. A staggering two-thirds of this net wealth is in real estate. Post-pandemic this has become even more extreme. Within real estate, most of this increase (three quarters) has come from price appreciated rather than investment (new homes or home improvements).
Before 2000, net worth largely tracked GDP at a global level. Since 2000, even as real investment continued to track GDP, net worth has surged. This coincides perfectly with the collapse in interest rates broadly and capitalisation rates on real estate specifically, which have played a decisive role in the tripling of real estate prices over the two decades. In the UK for example, McKinsey calculates that only 9% of the price appreciation in real estate was due to net capital investment.
The Importance of Leverage
For the vast majority, their only experience of the potential power of a leveraged investment is through their house, with ten to twenty turns of leverage de rigueur. From the perspective of most first-time buyers, the performance of this single leveraged asset, dominates the trajectory of their net wealth. We have already established that houses have done nothing but go up for decades. If we add to this the impact of leverage, the perpetual declines in the cost of leverage and the ability to refinance, we end up with the transformational impact on “net wealth” that the McKinsey report captures. A multi-decade compound 5.7% return is nice, a multi-decade leveraged compound 5.7% return is transformational.
However, those of us who’ve hung-out in financial markets for long enough are acutely aware that leverage is a double-edged sword. If house prices were to fall for a sustained period, those who bought recently may be on the wrong side of the biggest and most leveraged bet they will ever make. The societal impacts of such a shift would be challenging to say the least. Think of fewer wine tastings and more Molotov cocktails.
Timing
The only way house prices in the UK have been able to go from 3 times earnings to 7 times, is a combination of the amount that can be borrowed going up, and the cost of the borrowing going down. However, we know from our earlier maths that this cannot go on forever.
George Soros once said, “when I see a bubble, I rush to buy it!” The wisdom in this quip is that when a bubble is forming, leaning against it too early can be ruinous. How can we be sure the bubble has finally exhausted itself? In truth, we can’t. However, with the recent inflationary burst, and interest rate response means that the howling gale behind valuations is categorically over. Even if rates fall back in the years ahead, which is quite likely, the two-decade trend of falling rates is over. Low rates can stop prices from collapsing, but they can’t drive prices ahead of inflation indefinitely.
Psychology
The general population’s heuristics relating to the housing market have an overwhelmingly positive bias. Virtually all their lived experience of what house prices can do is based upon the last thirty years, which as we have illustrated, are the most unusual in history. Never, has humanity experienced such an enormous and protracted fall in rates, and the Bank of England has compiled data going back 5000 years to prove it. The vertical move at the right and edge of the Amsterdam chart above is all we have ever known. Shaking that level on entrenched bias is going to break some hearts, even if we know for certain that it cannot continue forever.
There is also a natural emotional attachment to housing that separates it from purely financial assets. I suspect Mrs Buffett wouldn’t trade the last 65 years on Charlie Munger’s couch for $6.6bn today (although I reckon Warren would).
Looking Forward
Much of the above focuses on what has happened, and the obvious limitations to extrapolating our recent lived experience in this asset class. Below, I indulge in some speculation about how the future may look.
As a starting point let’s assume the trend of recent decades continues for another twenty years. Let’s assume that house prices continue to increase by 5.7% with inflation at 3.6% and that wages increase in-line with inflation. Let’s also assume a 25-year repayment mortgage with a 10% deposit has an interest rate of 5%. Fast forward 20 years and the average house price has gone from £295k to £894k. The average salary is now £67k up from £33k today, taking the multiple of house prices to income to a cool 13.3 times. Assuming our new home buyer has managed to save 100% of his net salary for 2 years to scrape together the deposit, his monthly mortgage payment of £4,700 will (assuming tax rates don’t change) consume 108% of his net salary. I suppose this is one way to tackle the obesity crisis, but nevertheless, if think we can safely file this scenario under “impossible things”.
Let’s try a different scenario. Wages and inflation are 3.6% for the next 20 years, roughly doubling the general price level. House prices consolidate their recent gains and dribble lower by 1% or so a year, ending the period at 80% of current levels (£236k). The monthly mortgage payment for a new buyer on the same mortgage terms is now £1,240 or 26% of his net salary. Does this sound less impossible? I think so. The only wrinkle is this implies a 60% decline in the real value of housing. Broadly in line with what has happened time and time again throughout history and from much less extreme overvaluation. Given the broader anchoring of the public to recent history, I suspect many will find this highly plausible scenario distressing.
Embracing Willie Sutton
Willie Sutton was a famed US bank robber of the 1930s around whom an apocryphal story has grown up. When once asked by a reporter why he robbed banks he replied by saying “because that is where the money is.” If Willie had read the McKinsey report, he would have no doubts where the money was in 2023.
In the UK for example the value of residential property is around £8-9tn against which there is £1.6tn in mortgage debt. There can be no doubt “where the money is”.
Not only in the UK, but in virtually all developed market democracies, debt in all sectors of the economy is extremely high. Government debt levels are at all-time highs outside wartime and huge deficits are forecast as far as the eye can see. As we enter a new cold war, who has the fiscal capacity to double military spending and a percentage of GDP, provide state subsidies for strategic industries, transform the energy system and meet increasing entitlement commitments, all while paying rising interest costs on the debt? The answer is nobody.
As the pressure on the state to deliver on all these necessities grows, so will the desperation to finance them. The central bank’s printing press while undoubtedly be part of the solution but governments will be forced to be creative in funding themselves. Sooner or later, they are going to have to go where the money is.
In a recent opinion piece in the FT “The case for a land tax is overwhelming.”, Martin Wolf drawing on a paper called the Centre for Economic Policy Research in 2021 entitled “Post-Corona-Balanced-Budget Super-Stimulus: The Case for Shifting Taxes onto Land”, argues that land, as opposed to the capital stock created out of human endeavour, incomes and consumption, should be the focal point for tax policy. They argue that this would be a transformational positive for the UK. The problem is that in recent decade the credit system has morphed to almost exclusively service the most unproductive sector, housing.
Economists have long understood that it is sensible to tax factors of production, whose supply is unaffected by its price. The thrust of the argument is that there is a limit to how much you can or should tax incomes (either directly or through consumption) and productive assets without crushing prosperity. Socialising much of the yield on land is the obvious solution to this problem and the tax base is huge. And by definition, land can’t get up leave the country. The paper focuses on the land component, rather than the value attached to any buildings on it. They also concede that it would be a major negative for land (and by extension) property prices.
My point here is not to discuss the details but rather capture the main implication. That is that while taxing land and property may have been politically tricky in the past, at some point in the future it will become imperative to fund what must be funded and at the same time unburden the economy from the dead-weight of housing that it has struggled to carry for so long.
Summary
If youth is wasted on the young, then wealth is wasted on the rich. Historically, growth in net wealth was reflected in investments of the sort that drive productivity and growth. In the last twenty years, that growth has shifted almost entirely to unproductive property assets. Net wealth is simply a claim on future production. To the extent net wealth is stored in unproductive assets that cannot support future production, then that net wealth must, in the fullness of time, prove to be illusory in real terms. Converting today’s net wealth into assets better able their real value over time is the game at hand.
Houses are real assets over the very long-term but not from any starting price. They are inextricably linked to the general price level and wages.
That link has been distorted by a unique period of falling interest rates that cannot be repeated.
This has had a profound effect on net wealth and its composition, ultimately to the detriment of the real economy.
Recency bias causes people to extrapolate, even though this means believing impossible things.
Economic reality will mean in the years ahead, the risk of a shift to taxing land/property will become increasingly likely. This risk has not yet been priced in.
To believe that UK housing will deliver positive real returns going forward requires you to believe any number of impossible things. House prices don’t need to crash, and I don’t believe they will. If they simply retrace some of their recent gains over a couple of decades, the hit to real values will be material as the general price level eventually catches up and even overtakes.
Investors need to look to more productive investments to protect and grow their purchasing power.
"It takes all the running you can do, to keep in the same place. If you want to get somewhere else, you must run at least twice as fast as that!"
― Lewis Carroll
