Dear Reader,
This will be a four part series of fundamental analysis and valuation of the housing market.
Part 1: I will explore whether or not a housing bubble exists by estimating the intrinsic value of real estate and comparing it to the price.
Part 2: I will look at how large the bubble is, pontificate on how the bubble formed, and attempt to explain why it exists.
Part 3: I will illustrate the similarities between the housing market and the economics of a Ponzi scheme.
Part 4: I will look at whether or not the bubble will burst and what that might look like.
Although I valued residential real estate markets in 108 of the largest economies in the world, and found comparable results, I will focus my discussion on the UK.
Part 1: Is there a housing bubble?
As I established in a primer on valuation fundamentals, the intrinsic value of any financial asset is the discounted value of all expected future free-cash flows. For real estate, these cash-flows are proportional to all the rental income it could produce now and in the future and the cost of financing these assets.
Prevailing sentiment suggests that real estate prices always go up as we have an increasing population and a limited amount of land. I’m sure you, like me, have encountered this line of reasoning many times. It seems logical and easy to understand, but it completely misses the most fundamental question that is easy to ignore in a rising tide: If the intrinsic value of real estate is derived not from the present value of all future rents received, then what is it derived from? Outside of the utility of living in the house themselves, what possible value could it have for the owner, or any future owner, if it is never rented out? Although scarcity can, if paired with rising demand, make the price of something go up, it does not make something increase in value of its own accord.
Now, in order to estimate the intrinsic value of real estate, we need to measure the three pillars of valuation: cash-flows (rental income), growth (in these incomes) and the cost of risk (mortgage and deposit rates).
To estimate the cash-flows, we need to know household incomes and the proportion spent on rent. Since 1950, the average household in the UK has spent 30.78% of their after-tax income on rent and in 2019 this was 31.18%. As the average household income was £29,172.00 last year, we calculate that the average household spent £9,094.49 on rent. After adjusting for maintenance costs, depreciation, operating expenses, management fees, taxes, and additional reinvestment requirements, the average property would have produced £2,415.82 in after-tax free cash flows in 2019.
Next, in order to estimate the growth in these cash-flows, unless we assume that landlords can take an ever-growing slice of the income pie, we must assume that total landlord incomes must grow at the same rate as incomes in perpetuity. To do this, we note that since 1950 the prevailing local interest rate has been a reliable predictor of nominal income growth (a correlation of 0.89). Intuitively, income growth must be high enough to cover interest, lest labour become uncompetitive and all capital flows to debt holders; while income growth cannot far outpace interest, lest capital become unattractive. In any given year, interest rates and income growth are likely to differ slightly, but over the long-term they should converge. Thus, we will use the interest rate to estimate growth in household incomes. The prevailing interest rate at the time of writing was 0.31% and this is what we will use.
Finally, we must consider the cost of mortgage financing and the cost of equity. In the UK, the APRC on a 10-year fixed-rate mortgage was 2.74% and my estimated cost of equity for the marginal real estate investor was 3.45%. If we assume constant debt to capital ratios, then the cost of capital for the average real estate investment in the UK is 2.91%.
The final result of this is that the average property in the UK rented to the average household is worth £92,784.03 but is currently priced at £231,185.00. I ran these valuations for every year back to 1950 using the interest and mortgage rates, household incomes, the proportion spent on rent, taxes, and household debts at the time. The result shows that since the mid-80s, the gap between the intrinsic value of real estate and the price has been widening. As households earned more, the intrinsic value of real estate grew, but the price grew much faster, spurred on by an unprecedented expansion of the credit-supply. The financial busts of 1987 and 2008 did little to correct house prices back to their intrinsic levels. Houston, we have a bubble!
Until next time.
Please subscribe and share this newsletter.
As a subscriber, you will be notified as soon as new articles are released and will have access to specific subscriber-only content. This will soon include specific company and investment analysis.
Sorry to keep commenting but these posts are quite fascinating. This statement:
> Since 1950, the average household in the UK has spent 30.78% of their after-tax income on rent and in 2019 this was 31.18%.
I think this is what could allow the bubble to continue for a long time. Wages won't really grow in the UK that much in real terms but the % of income spent on rent in the short term can definitely still grow :(.
As seen here: https://www.money.co.uk/current-accounts/the-cost-of-rent-index
32.5% in the UK (London skews this higher). The top countries are around 50% of income going on rent. We still have some ways to go before this amount is spent on rent. There's obviously a hard cap to this before people start moving countries or just going homeless and bankrupt, however we are still probably away off from this hard cap.
Obviously this is terrible as it will mean a drag on the economy as more money goes from actual productive businesses and consumer goods and into rent.
But considering western countries approaches to continually propping up the housing market with shit like 'help to buy' and 'stamp duty holidays' which inflates prices (by helping the housing companies) I can see rent increasing as a % of income to be 50% in the next 20-30 years.
Now like you said this isn't sustainable for it to keep growing in perpetuity...
I hope this housing market does indeed pop or revert back to the mean soonish as it will be a drag on the economy.
> Next, in order to estimate the growth in these cash-flows, unless we assume that landlords can take an ever-growing slice of the income pie, we must assume that total landlord incomes must grow at the same rate as incomes in perpetuity. To do this, we note that since 1950 the prevailing local interest rate has been a reliable predictor of nominal income growth (a correlation of 0.89).
I don't really understand this part too much. Please can you point me to which website you got this data of local interest rates from?
> Thus, we will use the interest rate to estimate growth in household incomes. The prevailing interest rate at the time of writing was 0.31% and this is what we will use.
What happens if local interest rates rise later on, wouldn't we be underestimating the terminal revenue growth in this case?
I've inputted the same numbers and got a similar output as you in my UK Real Estate DCF. However it seems that the DCF is highly sensitive to small changes in Terminal WACC and Terminal Revenue (rent) growth. This is why I'm trying to understand this part especially well to make sure it's right.