Hacker News Reader

Comment by mastax | Hacker News Reader

That index measures the affordability for buyers. If the houses are very affordable for their current owners (due to low rates locked in), what's the mechanism for them to default? Mass unemployment, I suppose. But that doesn't seem causally linked to the affordability index.

View on HN

Falling home prices and default rates are absolutely correlated.

Again, just because we don't yet know the trigger which will cause this doesn't really matter.

Home affordability is mean-reverting and the only way for it to revert is waaay higher incomes or lower home prices. I'll bet on lower home prices.

> Home affordability is mean-reverting and the only way for it to revert is waaay higher incomes or lower home prices. I'll bet on lower home prices.

Yes and no. It is or can be mean-reverting in the short term but not in the long-term else we have a deflationary economy which would be a disaster and we'd do anything to prevent it.

> Again, just because we don't yet know the trigger which will cause this doesn't really matter.

I think you're starting from a conclusion and working your way backward. It's no different than "the market will crash again at some point we just don't know why". You need to identify the trigger and then draw the conclusion. Not the other way around.

BTW I see arguments for and against the general claim about mortgage defaults and it blowing up again.

For - housing prices are way to high, banks bought mortgage-backed securities with rates that are too low, layoffs mean defaults, etc.

Against - ongoing supply chain issues, housing supply is severely constrained everywhere in the country, low interest loans from those who keep their homes prevent them from moving, etc.

but this is all surface level. Need specifics to draw meaningful conclusions.

> Yes and no. It is or can be mean-reverting in the short term but not in the long-term else we have a deflationary economy which would be a disaster and we'd do anything to prevent it.

This doesn't make sense.

If affordability is always trending lower it means that housing is always getting more expensive in real terms. We know this isn't true.

> I think you're starting from a conclusion and working your way backward. It's no different than "the market will crash again at some point we just don't know why". You need to identify the trigger and then draw the conclusion. Not the other way around.

Interestingly, we can do that with many mean-reverting processes. Many, many people called the housing crisis on 2008 based solely on extreme valuations. Same with the market correction of the past year. Same with crypto correction.

You can count on extreme valuations eventually mean-reverting.

[edit]I think they are saying that housing prices may revert to a shorter term mean and not say a 50-100 year mean because if it did the latter it would mean that there was serious deflation in the economy.

Anyone that called the housing crisis in 2008 SOLELY on extreme valuations got very, very lucky.

Housing prices are not mean-reverting in and of themselves. None of this happens in a vacuum.

> None of this happens in a vacuum.

Doesn’t this hurt your case as well?

Every housing chart I’ve seen for metro areas shows a distinct boom/bust cycle in housing that is indicative of mean reversion to around 25-30% of payment to earnings ratio. Many metro areas in US are over 40% now.

I don't know how it would hurt my argument - I'm saying it's more complicated than 'prices are high so there must be a bubble'

Over what period? Have there been any fundamental changes in the financing or supply of housing? How about the people purchasing homes - are there any differences between cycles? It's also possible that income rises without housing prices falling at all. That hasn't happened since the 60's, but it's possible.

Also, I don't think there are distinct boom/bust cycles in housing generally, because the charts lack the necessary trough. Look at the House Price Index for New York, for example; it is not representative of a distinct boom/bust cycle at all; price retractions are a fraction of the previous growth and last for shorter periods over the last 47 years.

Further, I don't think it's indicative of mean reversion to around 25%-30% because the house price growth rate has outpaced the income growth rate by 2-3X for over 50 years.

Are you saying that this only applies to metro areas?

Could you share some examples of this data? Mine mostly comes from the St. Louis Fed: https://fred.stlouisfed.org/

The following chart shows clearly the boom and bust cycles since 1960 of US and UK markets and how the price/income ratio reverts to around 4.5 to 5. The current statistic is near an all time high of 7.5, 50% higher than it should be.

https://www.longtermtrends.net/home-price-median-annual-inco...

So what you refer to as the bust is when it reverts from boom to the mean, and your evidence is that this has happened...once?

This isn't really a chart showing a cycle, at least not a multi-year cycle. You have 40 years of between 4 and 5, then a 5 year bubble and a retrenchment that bottoms out at about 5 before rising again.

There's no consistent cycle, though. If, for example, every 20 years you had it cross 5 and go to 6 then go back to 4 for a few years before repeating, that would be a cycle.

Cycles are clearly shown in the chart. They do not have to be consistent in gap as no economic business cycles present that way. Of course, these numbers are aggregate across all housing in 2 countries so the cycles are muted versus local min/max demonstrated in specific metro areas.

Even small swings from 5 to 4 demonstrate a 25% swing in aggregate housing prices, which is still significant. The whipsawing recently since 2000 represent two swings of around 75% in aggregate housing prices relative to incomes.

> If the houses are very affordable for their current owners (due to low rates locked in), what's the mechanism for them to default?

Here’s one scenario …

1. Forced to move or choose to move, but keep house and rent it out 2. Housing prices drop along with rents that do not cover mortgage 3. Owner tires of losing money, but cannot sell house because underwater on low interest rate mortgage 4. Default

Hmm.. don't people tend to borrow against their houses? If housing prices fall 10-20%, people with poor risk management might get effectively margin called, forcing them to liquidate, which becomes a contagion event.

Mortgages are typically issued as non-recourse loans. This may not be true for HELOCs or other risky equity loans, but usually your lender can’t seize your home unless you haven’t been paying monthly.

Foreclosure takes many months.

I'm pretty sure HELOCs can be called any, iirc, but I'm not sure if fail to pay the call means they put a lien on your house or force you into foreclosure.

When people say "you should buy real estate as an investment because you can't live in your brokerage account" and then do crazy stuff with their equity that they wouldn't even dream of when investing (e.g. Boglehead investors who HELOC for some net negative remodel) it really blows my mind...

HELOCS cannot be called, but they can be frozen, preventing further borrowing, or have their borrowing limits reduced.

You're on the right track. There won't be a margin call, but deflating/stagnant values and high interest rates prevent these lines of credit from being established. This is what causes the overall economic slowdown in the US. People use these HELOCs for all sorts of spending they can't afford otherwise.

Reply on news.ycombinator.com