Hacker News new | past | comments | ask | show | jobs | submit login
This is the United States’ third biggest housing boom in the modern era (nytimes.com)
155 points by olivermarks 10 months ago | hide | past | web | favorite | 158 comments



I believe as long as there are people who make much more money than they can invest productively in non-housing they will keeping buying houses and drive prices up. That's one of the effects of income inequality. The people who are on the winning side make more money than they can invest.


There's no such thing as "more than you can invest". Investment in public stocks scales very well -- $100B is no problem.

Directly owning real estate, on the other hand, doesn't scale that well. You need to hire property managers, and deal with municipal tax and assessments for each property. If you own several properties, you have to deal with hurricanes and earthquakes and fires and local zoning officials more often than you'd like.


Owning raw realestate requres management, like any other business. but there are REIT. I'd much, much rather have 5 million in a vanguard REIT than Seattle/SF realestate...


Taking loan on a reit isa different issue, however.


It would be similar if they allowed people to use their ownership of stocks and index funds to use as leverage to borrow funds to buy more stocks. They could borrow based on the risk ratings of stocks, and everyone knows it's a fact that index funds are the safest investment out there so the risk on that should be close to 0 and allow the biggest leverage. I'm going to stop right there, I'm starting to frighten myself.



Index funds usually halved in value during the financial crisis. They still have exposure to whole-market crises, even if single sector or company crises don't touch them much.


I was being sarcastic there. Everyone also knew for a fact that property is a very safe investment and things backed by property is a very safe investment.

Before the stock market took a knock over the past two years, people were talking the same way about index funds as they were talking about property. The same thing with bitcoin really.


I have a bit less faith in real estate than you do. My impression is that, from a regulatory point of view, real estate is regarded as a safe investment because that results in cheap mortgages, and cheap mortgages are popular among voters.

It's not that real estate is safe per se.

Index funds usually recover to before-crisis levels in about 5 years, so they're not terribly risky in that sense.

The Bitcoin believers were always lunatics. That people will believe in stupid shit is nothing new.


People do that. Plus far more people do it 'indirectly' through options.


So people who are working under this public stock company will buy more house. Stock is not a physical investment, but indirectly converted to such.


That's... extremely indirect. When you buy a public stock, there is no flow of money from you to the company. That's stock they already sold to someone else, a long time ago.


I don’t think this is a great explanation. If this was driven by income inequality and investment dollars looking for investments we would expect the individual home ownership rate to fall as large investment funds bought up homes as investments. Instead the homeownership rate has been climbing for the last 2.5 years.

I believe the better explanation is the Federal government. Both in the form of extraordinary easy monetary policy and in the form of extraordinarily generous mortgage lending terms in its role as far and away the largest owner/guarantor of residential mortgages.


The thing is the property bubble is predominantly in cities. And those properties are much more rarely mortgaged - its kind of an all or nothing deal if you can outright buy a multi-million dollar condo or not, banks are much less willing to mortgage one.

And there the prices are only going up due to NIMBYism about building for density. There is sea change in our cultural dynamics that is following in the wake of widespread adoption of technology and the Internet - we aren't isolating ourselves on quarter acre copy pasted plots an hour long commute by car from the office, we want to live in 500 square feet of space a few miles by bike or one station away from our office. We marry less, we have less kids, but there is absolutely nobody building anything for this demographic.

In an even more broader context we should be incentivizing people to live as close to peak urbanization because its the optimization of the cost to house and maintain people. People are cheaper for all of society and for nature abroad if they live closer in denser housing with more public infrastructure to support them than to have them living in repeated stick and glue mcmansions on HOA mandated lawns that require a thousand liters of water a year to keep green in the midwest.

We are literally doing the opposite of what we should, and its going to cost economic growth for our entire lives in all likelihood.


My experience in Seattle is different. Almost every house or condo in the east side or Seattle is a million plus, and it's really common to put 500k toward one, and pay a substantial mortgage (more than half a mil). So these can be lost with reduction in value combined with a short term mortgage that needs to refinanced, like happened in 2008. If you are a software engineering with 5-10 years experience and a family looking for a house you are in this game probably. These people have a good job but they don't need and often don't have money to cover the rest of their house.

I agree with the comment about lack of density in housing leads to these prices (cause people keep freaking moving here). And the point about not much housing being built for people leaving in smaller apartments like 500 sq ft in the dense area applies here. There seems to be mostly refurbished old housing in that range. I guess developers don't think there's a market for that (single person?) or they can't make money or both.

I think we need all these kinds of housing (higher density, also smaller places closer to work and transit). People don't priotize transit enough in purchases I think, but of course you can only buy what's built in condos.


From the data I’ve reviewed, this is generationally biased. Older people like buying houses. Younger people, when one restricts the population to those who can afford it, don’t show as strong of a bias.


Maybe because the young people are facing a higher housing prices, and they don't believe that it could raise for the same amount in the future.


Maybe Young people are more mobile. The security of buying a house is beneficial when you’ve got kids at school. When you don’t mind where you live as long as there WiFi and transport renting is beneficial, you can move across country on a whim.


If you require public transit, there are only a handful of places you can live, each with unique and volatile conditions that make owning especially attractive. It’s suburban sprawl which is almost perfectly fungible nationwide.


Depending on the type of transit, owning near the stop is only attractive until the transit operator eliminates your stop or your route.


>Maybe Young people are more mobile.

That wouldn't explain why young people in previous eras bought houses sooner.


> That wouldn't explain why young people in previous eras bought houses sooner

It could. Moving across the country is cheaper and less risky now than ever before. That certainly factors into my preferences around home ownership.


>It could. Moving across the country is cheaper and less risky now than ever before.

Is it? Rent itself was much cheaper in the past, even in New York.

As for the risk, back in the 50s to 80s, a worker had many options of finding work in tons of places. Now, much less so.


Maybe it's also a value thing, where the young people nowadays have much more liberal values in general. Also technological advancements like the Internet, etc.


Well, young & old in this context roughly translates to millennials & baby boomers.

It is the former that are more likely to have kids in school, not the latter.


What is the source? A quick search is not finding it.


Older people have access to cheap capital, and many younger are nowhere near being allowed a loan.


Or maybe younger people can't afford homes and the data is showing that, not bias.

This is certainly the case with me, and absolutely every other person I know in their 20s/30s who isn't in a upper-management position at a large company. I'm currently making above the median family income in my country and have no hope of owning a home anywhere near I can get work.

Meanwhile my old co-worker that I used to work with at Safeway as a teen bought a house in the 70s working as a shelf-stocker. In fact both of my co-workers at the time were 40-60 and bought a house working part-time as shelf-stockers early in life. You think a shelf-stocker at Safeway has any hope of buying a house today?


In some regions, yes. Likely not in urban regions.


Or, you know, they can't afford buying a house, and make a virtue out of necessity.


There are always investment opportunities outside of housing. Just look at Warren Buffett and Charlie Munger.


Buffett has been candid about not being able to find good investment opportunities lately... Berkshire is sitting on a mountain of dry powder (in short term Treasuries) and iirc has (shockingly for them) considered buybacks or dividends because they say everything is overpriced.


That's completely different. They are waiting for the best possible deals during an economic downtown -- something that they have always done.

They will continue to buy stocks.


Why do a lot of the millionaires and billionaires buy overpriced housing in London, Vancouver and other places instead of investing in something else? Speaking about too much money: Companies like Apple sit on hundreds of billions of dollars and don't know what to do with it. Theoretically they should invest it in new profitable products but they don't.


Because it can be safer to park their money in countries with strong private land ownership laws than keep it in their own country, even if the home drops in price a bit i.e. Chinese buyers.


I don't understand why this comment was downvoted--my understanding has been that a major component of rising housing prices globally is Chinese investors wanting to move their money out of the Chinese economy and into something they perceive as more stable, less corrupt etc.


> Chinese investors wanting to move their money out of the Chinese economy and into something they perceive as more stable, less corrupt etc.

That's an extremely favorable reading of the Chinese(Russian, African, South American, etc.) buyers and of the markets they're buying into. These are often, and I would argue largely, the corrupt connected elites who dominate corrupt economies trying to move money out of their countries before losing power, being caught, or before disasters in their local economies cause their investors to attempt to cash out. They are helped by a west that has lost manufacturing capacity and largely survives economically by circular credit scams that need a constant influx of these colonial resources to maintain the illusion of growth.

The best thing about real estate is that it never goes to zero; the best thing about western real estate is that your networks of accounts and holding companies will never be seriously investigated, because skimming from and taking commissions from transactions between those things is the only inflow of wealth, source of new investments, and source of political contributions. You can also let your kids live in it while they're going to school in London and NY.


> Why do a lot of the millionaires and billionaires buy overpriced housing in London, Vancouver and other places instead of investing in something else?

Having a big house is nice. Living in a major city is nice. The combination of the two is very nice, but expensive.


The whole point of questioning "millionaires and billionaires buy overpriced housing in London, Vancouver and other places" is that it's not about living there - overwhelmingly, they just buy housing. They don't live in these places, they don't intend to live in these places not even when visiting that city, they either get rented out but often simply stay empty. In the top end housing of e.g. London the prices get set by people for whom it's purely an investment, not about having a fancy home.


Worse than that. They sit on hundreds of billions of dollars but still borrow money in the tens of billions thanks to fed's still very low interest rate. If that's not systemic issue I don't know what is.


Sitting on money in foreign accounts and borrowing US dollars to pay dividends was to avoid incuring tax on repatriation. That shouldn't need to happen anymore since taxation on foreign income is no longer deferrable by delaying repatriation.

I still expect companies with lots of cash to borrow for some expenses and anyway keep large lines of credit open, but it would be strange to borrow to pay a divident if you already have more than enough cash on hand.


Real estate is an excellent way to launder or park money.


Some of it is really that they have an unbelievable amount of money and it's hard to spend it. Even really nice cars tend to max out around a million, and if you have billions that's nothing. Hence why you see people spending hundreds of millions on yachts that they rarely use, too. It's a good way to spend (and be seen spending) vast sums of wealth.

So too are really luxurious apartments in the world's elite cities.


The super rich don't really care much about the cost of the house they live in. They just pay what they have to pay to get the house they want. Overall this cost represents a small percentage of their net worth. That's why they don't care if it's over priced or not.

Any other house in any other market where they don't personally want to live get's a different assessment.


Can/do any of these companies with excess cash invest in the housing market perhaps through funds like REIT (for better liquidity than direct purchases)?

It looks like a good diversification from bonds if chosen judiciously but there could be regulatory reasons not to do it.


Where else would they buy? Agglomeration economics are real. No one is going to make Merthyr Tydfil great again.


They've already made it — they care about stability more than growth.


But we are told if we just give them enough tax cuts they will invest and "create" jobs.


As a non native speaker, a "housing boom" sounds like a boom in number of houses, not a boom in the price of existing houses :)


The two are closely correlated in practice. The boom in prices incentivizes developers to build more, leading to a boom in number of houses, which eventually leads to a bust. So the pendulum swings again. Control engineers would describe it all as a system that's not very well damped.


>leading to a boom in number of houses

The requisite for this seems to be little or no govt intervention. In Akl, NZ, for example, you're only seeing high-end homes being built yet prices & demand keeps going up for all housing. Now they're subsidizing projects for entry-level homes as if building a set number of homes (and failing to meet build targets by a long shot)will solve market woes.


There are zero problems with building mostly high-end homes. Then the rich can live there and the less rich can live where the rich previously lived -- win for everybody.

One of the best things the government can do in those cases is ensure that large housing developments are still zoned to allow them to be subdivided. Nothing wrong with four families living in one big mansion.


No, that's not right. First, there might not be enough demand for the most expensive places, only so many people can buy a 20 million dollar apartment. There's no magic power law that means that supply of housing works in a way that someone stepping up always makes a space for someone below. Also, the cost of a top end place could often pay for many many more lower end housing. Generally it seems easier to make a profit on expensive housing or whatever, because cost matters less in the final profit (that takes lots of discipline to make profits on more units of cheaper things generally).


> only so many people can buy a 20 million dollar apartment

Yes, and that's what turns the 20 million dollar apartment into a 10 million dollar apartment. If prices are too high, the market will correct for that by having apartment complexes needing to rent out units for less than they cost to build. That is why market research is important and should be on the developer to determine what there is demand for.


I heard the market corrected in Spain by allowing multiple generation loans https://www.dailymail.co.uk/news/article-2537511/The-40-year...


Does this get thrown off if the rich don’t buy the new high end homes to live in?


If you’re a developer, build a bunch of high end housing, and try to sell it off, and no one wants to buy it, you have approximately three choices:

1. Leave them empty to deteriorate. 2. Cut your losses and sell them for cheap 3. Let them out for less than you feel they ought to be worth in the hope that people will be willing to pay your high price soon.

Option 1. Is unlikely to happen as it is very unprofitable. This leaves options 2,3, each of which leads to people living in the newly built properties, so the total supply of housing has increased. Even if people are just renting, this means there is less demand for rentals so less of an incentive to buy houses to let them out, decreasing the demand for houses.


Developers rarely build blindly. If they build high-end housing, there is some demand which will definitely buy-in... and that opens a new option:

4. Wait, because operational expenses are cheap compared to the capital expenditure.

They will set an asking lease that is guaranteed to attract enough high-rent leasers to pay for maintenance + loan interest (if any), then hold the price right there regardless of vacancy rates. Rates will be lower only if the market demographics irreversibly change, but otherwise they can afford to wait as the building slowly fills over a decade or more - each additional renter is pure profit, and raises the value of the building as an asset for resale.

The alternative is lowering prices, which attracts a different demographic that makes it challenging to raise rates afterwards. They may generate more short-term income but overall value is lower.


> 4. Wait, because operational expenses are cheap compared to the capital expenditure.

The problem with waiting is that you're sitting on a huge risk, because holding prices artificially high attracts new construction. If that price makes construction profitable then construction will continue to be profitable until prices come down. To keep prices from coming down, the existing speculators would have to buy all of the new construction. The vacancy rate would keep increasing until it could no longer cover the maintenance costs.

And the higher the vacancy rate gets, the greater the downside risk to holding vacant units, because there are more competing vacant units that the other owners could at any point decide to put on the market, causing lower prices that induce other owners to abandon the conspiracy and put their units on the market too. And the more vacant units there are, the more the price declines when the cartel dissolves.


> Does this get thrown off if the rich don’t buy the new high end homes to live in?

Not really, it just implies that you should build even more of them. Soak up all the demand from rich speculating idiots, then let them take a haircut when it turns out you can build more housing than speculators have capital to buy with. As soon as they realize that prices are no longer going up, the speculation ends and normal people can go live in all the housing you've been building.

It's not as if building a different kind of housing would change anything. If you built two smaller units instead of one bigger one, you don't think speculators would just buy both of them with the same amount of money? What you need is to get the supply to the point that it's clear speculation will not be profitable.

There are also some known methods to discourage them, like vacancy taxes, if you're looking for a more immediate effect.


That doesn't happen in practice.


Sounds like San Francisco. A decent amount of new construction recently but it’s all ludicrously expensive. Also there’s some small set asides for low income folks, but not nearly enough. And anyone in the middle is especially out of luck.


SF has to deal with zoning restrictions. Between rapid immigration, the tech wealth boom, and restrictions on lot size/height/residency - obviously all the new constructions will be ludicrously expensive.

"Affordable housing units" is so double speak. A few token units are subsidized so council members can say they're fighting for the general public, pulling the wool over lower/middle-class voters as they're slowly displaced.

"Preserve our neighborhood character" (NIMBYism) is the other corrupt argument. The only thing they preserve is the rising value of their homes. Inevitably, the neighborhood is either white washed and rebuilt as wealth takes over or Manhattanized in a rush if the original residents wake up in time. People are the neighborhood character, maintaining cost of living is the only way to really maintain character.

The middle path of proactive growth, raising density restrictions to avoid housing inflation, is difficult. People move to neighborhoods because they like it as-is... no one wants change until they are impacted, but by then they have no say in the results.


As a native speaker, a "housing boom" sounds like a boom in the number of houses, not a boom in the price of existing houses :)


Yeah, “housing boom” sounds like increased housing construction whereas “housing bubble” is the term I’m used to for higher prices.


Swathes of unsold homes in places like Spain and Florida after the last bust suggest it can be both!


That's still a far cry from 2.2 billion vacant square metres that China has, enough to rehouse the whole of California.


It sounds enormous until you consider their population size and urbanization rate. At the rate of 0.75% per year, it equals to 10.5 million people a year. Assuming 40 square metres per person, the stock only lasts about 5 years.

In addition, real estate is the preferred investment vehicle among the older and on average wealthier segment of the population.

"Urbanisation in China increased in speed following the initiation of the reform and opening policy. By the end of 2017, 59.4% of the total population lived in urban areas, a dramatic increase from 26% in 1990."

https://en.m.wikipedia.org/wiki/Urbanization_in_China


How many people can soak up that high priced unused real estate? Not that many. I expect there to eventually be a housing and infrastructure crash in China, but maybe they can absorb it because they are so large.


One also needs to consider saving rate (very high) and mortgage policy to answer that question.


Is the housing stock growing or shrinking or staying the same?


As a native speaker, I agree.


Everything about our current economy looks like one enormous bubble about to pop, and I think it's going to be much worse than 2008, depending on what actions the government takes. I see malinvestment everywhere, from venture capital to the recent machine learning and cryptocurrency trends.

Edit: As a young person, it scares me.


I agree completely! The amount of misallocated capital and the rate with which it is flowing toward more productive uses is extremely disturbing.

There should be ten or a hundred well capitalized Tesla competitors by now rather than all this idiotic old money chasing (fearfully) after dying economies ...

It’s very disturbing to think how actively harmful these misallocations are — the amount of money that still thinks an investment strategy that aims to protect the value of existing fossil fuel investment is a good use of capital ... pure blind delusion.

We are well past the point where you can make more money by investing to extend the timeline of fossil fuel returns than you could make by accelerating yourself away from your fossil fuel investments ... somebody needs to remind the investment community about the value of skating to where the puck is going to be ...


The misallocation of capital is a natural result of capitalism, which is a distributed, greedy algorithm.


Trust your instincts on this. It will probably lead you to investments counter to the mainstream. When I was young I felt much the same way. Result: Multiple paid-for properties and a respectable retirement fund. I never participated in any of the spectacular gains of Apple, Yahoo, Amazon, Bitcoin and so forth. Never bothered me a bit. This approach also let me get bargains when people were leaving markets.


Where are you putting your money now?


Starting a new business


When central banks print money it destroys the connection of money and wealth. It degrades the signal.

Personally I love Bitcoin bc it protects against this. Sure, it may be falling while the Fed tightens, but the Fed prints more often than it tightens.


But bitcoins and other cryptocurrencies have fallen far more than dollars have in any short timeframe. The best answer is diversify, stocks in different areas, bonds, cash, maybe real estate.


Well sure, because there are fewer dollars. They're being decreased and hence it takes less dollars to purchase bitcoin. Yes indeed re: diversification.


It's already collapsing in the SF Bay Area. "PRICE REDUCED" signs are appearing. The bubble is popping. One real estate investor and one Realtor I know admit that prices have been dropping for most of 2018.


The data does not reflect this. SF and the surrounding areas are mostly still rising...monthly is still going up and the yearly is up maybe 5%. Most other parts of California are weakening though in price on a monthly basis.

https://www.car.org/marketdata/data/countysalesactivity


It’s obvious things cooled off around August specifically during the tariff changes. If these trade talks get worse I imagine the market cooling even more.


other things applying downward pressure to high priced RE markets nationwide: the new tax law that reduces a taxpayer's savings from mortgage and property tax deductions, and rising mortgage interest rates


I don’t think SF area prices will ever “collapse” - it will just stabilize or dip. Even if we experienced a heavy recession there are too few homes for the number of people who live here. And a recession would only deter new construction.


Time will tell


> The bubble is popping.

... or just deflating a tad before it stabilizes and waits for the next big inflation.


I wish I shared your optimism.


I think of it as pessimism in the sense that SF real estate will never ever be affordable to the middle class buyer


Please tell my landlord this.


Unfortunately, idealized market theories notwithstanding, rent goes up in recessions as well as upswings and independent of vacancy rates.

https://www.census.gov/housing/hvs/files/currenthvspress.pdf


That's because we still don't build enough. Watch as SB 50 dramatically changes rents in CA.


I hope so with all my heart.

But for example, this[0] (triangular!) lot is 1.6 million. So developed or not, low rents seem unlikely even in that neighborhood. At the risk of heresy, dare I ask if real-estate markets as they currently work have failed their communities?

[0] https://www.google.com/maps/@37.8134181,-122.2745241,3a,75y,...


I wouldn't draw any conclusions about the US housing market from SF Bay data lol.


This is city- and neighborhood- dependent. San Francisco as a whole is seeing 8% YoY gains [1], while Menlo Park has seen 0% median sales price growth [2]. Nearby Redwood City is seeing 10% growth [3].

The rate of increase is lower than in 2017, but most places in the SF Bay Area are still experiencing high growth rates. Those rates are apparently more in line with historical norms rather than the white-hot multiple-cash-offers waived-contingencies heyday of 2016 and 2017.

I don't doubt that there are many price drops. Realtors expecting offers 30% over ask likely were setting list prices with too much growth expectation and had to reset to the comparables' sales price. Last year using the current comps' sales prices would have been listing the home too low.

1. https://www.trulia.com/real_estate/San_Francisco-California/

2. https://www.trulia.com/real_estate/Menlo_Park-California/

3. https://www.trulia.com/real_estate/Redwood_City-California/


Seattle homes have reduced their prices, but sellers were trying to max out whatever they could get. The market has done a self-correction, but it is still a sellers market. Good homes still go for top dollar and go quickly.


Case Shiller index is the highest on record at the moment https://fred.stlouisfed.org/series/csushpinsa


In a normally functioning housing market (with a reasonable amount of inflation), wouldn’t this be the norm?


The index is based upon median income and median home sale price, not upon house prices as an absolute value. It’s an indicator of affordability. I don’t know if indexes are comparable across different countries and what they would mean.


No. The author here is cherry picking a timeframe [2012-2018]. Nice cherry pick. Unless you were smart enough to buy in 2012 that is gone: forever.

Home prices have barley hit their 2008 prices. These are prices ten years ago. If a stock hit its price from ten years ago I wouldn't consider it a "boom".

If you are a zillow user, look around at normal cities [Not manhattan,SF, or LA] linked below. If you have a minute, look at "Price / Tax History", and see what they sold at 2006/7/8. I see this pattern all over the US. I could add 20 more if you need.

To me we are almost back to normal. Sure this is anecdotal but, averaging all US home prices with a cherry pick is just as silly.

https://www.zillow.com/homedetails/5435-White-Heron-Pl-Ovied...

https://www.zillow.com/homedetails/6241-Gobernadores-Ln-Carm...

https://www.zillow.com/homedetails/6600-Sutter-Ave-Carmichae...


I'm curious what pattern you're trying to illustrate with the '06-'08 home values compared to now, and why 10 years ago is something you consider "normal". The market crash showed that the valuation of homes from '06 - '08 was anything but "normal". And Zillow data definitely shows that median home values are quickly approaching 2008 levels, with a more natural and stable climb this time around.


Do you have a reason to consider 2008 as normal?


normal: %3-%5 yoy increases, basically a little under the mortgage rate. 2008 wasn’t normal that isn’t what I meant.


Yeah, I took your comment to mean something like "2008 wasn't normal. But imagine a "normal" price in 2008, and how it would have risen normally over ten years. Then maybe it would reach the actual 2008 price, which is about what we have in 2018."

Not sure if that's correct, but sounds plausible.


If housing is overvalued, mean returns lower than inflation would be better, though admittedly not "normal".


> Home prices have barley hit their 2008 prices. These are prices ten years ago.

The total inflation over the last ten years has been ~20%, so you would expect prices of a stable asset with the same intrinsic value to be ~20% higher. But the pre-2008 prices were a known bubble -- overvalued by significantly more than 20%.

What you're saying is that it currently isn't as bad as 2008, which is likely true without proving that housing isn't still overvalued now.

And if the prices in Florida are now overvalued somewhat, what does that tell you about the prices in SF?


It is already popping... like it did last time and like it will do next time. Welcome to the financialization of everything!

Look here for more data: http://www.doctorhousingbubble.com/


I wonder if the application of AI to finance would effectively smooth out bubble/bust scenarios. If you really think about it a bubble is when assets are overvalued usually due to herd mentality and fear of missing out. If AI we're more involved it would be able to short sell when something was overvalued and buy when it is undervalued. This is the essesnce of good investment and you would think machine learning would be able to gather much more raw information than any person and come to these conclusions.

I think this is what Renaissance Investments does to make such amazing returns.


Merely having a correct investment thesis isn't enough to make money - you have to also get the trade timing and execution correct, and bubbles have a nasty tendency to stay inflated long enough to squeeze out the shorts.

Like, the more profitable plays are often the ones that contribute to the boom/bust cycle. In the 2008 crisis, the big winners were people who took out as much mortgage debt as possible and aggressively did cash-out refinancing, ending up strategically defaulting on houses they got more money out of than they put in. When you're in a bubble, people tend to assume that the worst-case scenarios won't happen, so they'll extend credit that you cannot pay out in those scenarios, and you wind up getting some free expected value and positive skew that way.


Now that's a depressing perspective.


I'd suspect that a major difficulty of applying AI to real estate is limited fungibility and liquidity.

With a stock or currency, there's usually enough of a live market that you can sell quickly at roughly market prices. Much of this is due to fungibility-- one share of INTC is worth the same as any other, so they can trade sight unseen with narrow spreads.

With real estate, the AI dashboard can flash and beep and scream "Sell the houses in Roanoke now!" but there's no easy way out. You're either spending weeks or months closing out individual retail sales, and if anything goes wrong you're back to square one while the price is deflating around you, or you try to sell it commercially sight-unseen, for which you're going to take a major price haircut.

Might work somewhat better with commercial real-estate which seems a bit more oriented to grading and classification to allow more sight-unseen trading.


AI wouldn't smooth out bubble because bubbles are opportunities for a few to make wealth at the expense of others.

Bubble in America (at least the recent ones since Greenspan) are all down to Fed being asked to bring the society to full employment (which is so backwards as a law because a central bank cannot bring full employment).

The bubbles also get created because the real risk takers (big investment funds, banks, companies) are not allowed to take losses that the masses will take.

As an example, let GE, GM, hedge funds, banks who lent outsized mortgages take the loss this time while interest rates rise to "normal". Sure a few people will lose jobs but we've created this hazardous situation right now where morally bankrupt corporations are backstopped by tax payers because they're too big to fail while the individual will lose their house if they can't pay mortgages.

Companies have to be allowed to fail. Fed needs to maintain price stability on CPI and assets (which they couldn't after the last bubble).


Are you suggesting individuals should be able to default on the mortgage but still get to keep the house? Or that we should let the mortgages foreclose and people get kicked out, but not perform quantitative easing in that recession?


I'm suggesting taking capitalism to some true form. Let badly managed companies fail, let badly managed houses of individuals get foreclosed.

For innocent individuals, create a social security net. Instead of bailing out companies or keeping interest rates lower (as all the rich are asking right now so that stock market bubble stays up), use that same money to bail out underwater individuals in some way.

Right now, the whole market is a perverted system saving the rich every time. That is not healthy at all.


> Instead of bailing out companies or keeping interest rates lower (as all the rich are asking right now so that stock market bubble stays up), use that same money to bail out underwater individuals in some way.

Keeping interest rates lower was how they bailed out underwater individuals. Lower interest rates stimulate demand for housing, shoring up housing prices for the people who were underwater. Of course, that reinflated the housing bubble, so here we are again.

What they need to do now is to slowly raise interest rates. Raising them quickly would put us right back to 2008. Raising them slowly lets housing prices decline over a period of years at rate that doesn't instantly put everyone deep underwater again, or kick every adjustable rate mortgage payment above the level that people start defaulting again.


> Keeping interest rates lower was how they bailed out underwater individuals. Lower interest rates stimulate demand for housing, shoring up housing prices for the people who were underwater. Of course, that reinflated the housing bubble, so here we are again.

No. I meant using TARP money to wipe out a portion of that debt or create jobs instead of QE. The banks did not need bailing out. Certainly the execs shouldve been arrested.

Because of decisions last time around, the execs take risks with impunity now.


> Certainly the execs shouldve been arrested.

Some of them were. What really should've happened is that the big banks should have been broken up, and we should've fixed the regulatory environment to make it more viable to operate a small independent bank. But it's never too late for that.

> I meant using TARP money to wipe out a portion of that debt or create jobs instead of QE.

TARP "cost" -$15B. The government turned a profit. If you wanted to cancel TARP so the money could be used for something else, the other program would have to somehow also cost -$15B. If you had such a program we presumably should have done both and made $30B.

Moreover, the total US mortgage debt is ~$8.8T. The only thing that's going to make much of a dent in that is inflation over time -- which is what QE does. It also lowers interest rates, meaning that homeowners are paying less interest on their existing mortgages. It also lowers interest rates for government debt, which allows the government to fund more programs with less tax revenue. Ever wonder how Obama managed to pass the ACA subsidies while passing a tax cut? QE. Allows the government to pay less interest on the debt and use the money for something else.

QE doesn't cost money. It creates money. Far more money than the government profit from TARP. The "cost" is in inflation -- assuming you weren't otherwise fighting deflationary effects to begin with, which we have been and still are.

"Unwinding" it -- meaning getting the level of consumer debt back down to a reasonable level -- is probably going to require more. What you need is for people to have an incentive to reduce debt, also known as raising interest rates (as has been done), but then you need something to counter the deflationary effects of doing that. This is why interest rates are normally raised in a booming economy -- booms naturally cause an increase in borrowing and therefore inflation. But this "boom" hasn't seen much inflation because of various countervailing deflationary effects (housing bubble wanting to deflate, software eating the world, much of the boom being caused by low interest rates to begin with, etc.), so that isn't there, which is why interest rates remained at zero for longer than they probably should have.

The best thing they could do right now is to have another round or three of QE, but use the money to fund a UBI rather than to lower bank loan interest rates, then let the higher loan interest rates cause people to pay down their mortgages (using the money from the UBI) and deflate the housing bubble over time without causing an immediate crash or catastrophic general deflation.


I see your point. But I think if we have more rounds of QE and use the money for UBI, we'll get to Japan levels of disenchantment for 20 years (or perhaps more because we have far more debt to repay).

What I would like to see is some bad actors really getting wiped out, our GDP reduce a bit, asset values go down. That will cost a few jobs in the beginning.

But, in some time deflation will reach a plateau (because everyone needs to buy food, but that's not as low as we need to get).

Once plateaued, it would allow savers and risk takers to start investing in businesses themselves (these are the guys today with non-speculative money but unable to invest in overpriced assets).

Ultimately, bad actors from 80s, 90s and 00s are still alive and still investing right now. If the moral hazard is always solved by inflation, guess how many more speculative bubbles we will have.


> But I think if we have more rounds of QE and use the money for UBI, we'll get to Japan levels of disenchantment for 20 years (or perhaps more because we have far more debt to repay).

What happened in Japan was gradual deflation -- certainly something we want to avoid, but the opposite of what creating money by fiat does.

The problem we currently have is that some assets (especially housing) are overvalued. There are two quick ways out of it and a gradation of slow ones.

The first is that we just force a housing crash by instantly setting interest rates above 5% and suffer the consequences. This is probably the worst option. Its primary virtue is that it revalues housing and interest rates to more realistic numbers right now. Its short-term costs are very high.

The second option is better but still not great, which is that we do the first one and at the same time instantaneously create a few trillion dollars in cash by fiat and hand it out, so that housing prices crash in real dollars but stay the same in nominal dollars, and the price of everything else (including wages) has a big one-time jump to catch up. Then nobody is underwater on their mortgage, nobody defaults, banks don't fail, etc. But it would still be very disruptive and screw over one side of every fixed-price contract on anything, require all prices to immediately be renegotiated, etc.

The third option is to do the second option in slow motion. Eliminate the mortgage interest tax deduction, slowly raise interest rates, generally encourage people to take out smaller mortgages, all while creating new money (and transferring it to citizens without obligation) to offset 1:1 the destruction from the net debt reduction. The faster you want this to happen, the higher an annual rate of inflation you need. It may or may not cause nominal housing prices to actually go down depending on the pace, but real housing prices would decline until they're reasonable again. It's much less disruptive but takes longer.

This is still the best option even if you demand quick results, because even doing it at an accelerated rate could complete the task in less time than it would take for everything to recover from the massive shock of trying to do it instantaneously. In other words, if it would take five or ten years to recover from doing it all at once then it's less disruptive to do it gradually over the course of the five or ten years it will take regardless.

> What I would like to see is some bad actors really getting wiped out, our GDP reduce a bit, asset values go down. That will cost a few jobs in the beginning.

The difficulty is in restricting the damage to bad actors. It's obviously undesirable to wipe out everyone who bought a home at inflated prices just because they had to live somewhere, or the retirement accounts of everyone whose company IRA only gives them a choice between mutual funds that all have similar exposure.

> But, in some time deflation will reach a plateau (because everyone needs to buy food, but that's not as low as we need to get).

The problem with deflation is that it's a destructive feedback loop.

First people have less disposable income, e.g. because interest rates are higher and they have to spend their salaries on interest rather than consumption. Then they reduce optional consumption, and those industries downsize and people lose their jobs. Then those people stop buying everything except necessities because they have no jobs, so more other people lose their jobs. With unemployment on the rise, wages fall. With lower wages, there is even less consumption, so more people lose their jobs or have to take lower wages, and so on.

Meanwhile investors notice that "holding cash" is suddenly a profitable low-risk investment strategy, so who needs to invest in medium-risk enterprises that generate value and create jobs? Not only does the stock market crash (along with everyone's retirement accounts), normal job creation ceases and market failures become abundant because no one can raise money to start a business or enter a market. The worse the markets do, the more people want to hold cash instead and the worse the markets do.

Meanwhile the real value of all existing debt increases, forcing people to default, and defaults accelerate deflation.

General deflation is utterly catastrophic.

If you want to raise interest rates, either there has to be significant inflation already happening to offset the deflationary effect, or you need to offset it with a countervailing force, i.e. print as much money as it will destroy.

> Once plateaued, it would allow savers and risk takers to start investing in businesses themselves (these are the guys today with non-speculative money but unable to invest in overpriced assets).

You don't need deflation for this. The thing that causes the high asset values is really the low interest rates -- it's the same as the housing market. If risk-adjusted returns are higher than interest rates, people will borrow money and use it to bid up returns-yielding assets until they're not.

If you want higher returns you need higher interest rates, but raising interest rates doesn't have to cause deflation as long as the money the higher interest rates destroy is recreated and used for something else -- like a UBI. Or even just tax reductions and the like.

> Ultimately, bad actors from 80s, 90s and 00s are still alive and still investing right now. If the moral hazard is always solved by inflation, guess how many more speculative bubbles we will have.

Deflating bubbles through inflation doesn't actually reward the people who pay too much, because the mechanism of operation is that the nominal price of the overvalued thing stays about the same while the price of everything else increases. The people holding the overvalued asset pay by having negative real returns, even if the nominal returns are zero or slightly positive.


I agree with all of your thoughtful response and agree that option 3 is the best.

While a deflationary spiral seems bad, and I agree its a bad spiral to go down, a deflation after massive inflation is ok. A deflation in asset prices through rising interest rates frees up surplus capital in the hands of people.

For example, lesser rent/mortgage payments means more cash in hand to spend on say furniture.

Lowering of prices would also allow other investors to take some risks. For example, if the cost of input goods into a factory start reducing, that would free up capital to hire more people or do more R&D.

Of course, that only works while deleveraging from a bubble. I agree that deflation from "median" yoy inflation is probably not healthy.

Too much deflation will also cause people to lose jobs because of lack of investment. But like I said, there is a floor to it. Capital freed up because of lower asset prices will circulate in the economy again.

At the essence of this discussion, we are picking winners and losers. I'm not a fan of government making those picks. But, from what we've seen over the last 25 years, impunity from risks has created moral hazards already (such as promising 7% yoy growth to pension funds) with no concern of risk. The lenders need to take some risk and only that way will speculation end.

For example, without bitcoin bust this year, people would've kept speculating the value of bitcoin to the moon (as was happening last year). With this bust, bitcoin denominated assets are still denominated the same but on real terms, they are deflated.

Will that kill the bitcoin industry? Maybe. But for sure it will get rid of the excesses, speculation, fraud, overpromising and basically make humans aware that "Investments are subject to market risk. Please read the offer document carefully before investing"

To conclude, while defaults sound bad, they're not as bad after such a bubble is created (so long as they don't cause systemic faults, at which point, the system should switch to minor inflation, UBI)


> For example, lesser rent/mortgage payments means more cash in hand to spend on say furniture.

The mortgage payments don't really go down, because of the higher interest rates. You pay less in principal but more in interest. It also encourages people to take out shorter mortgages with higher payments so they can stop paying high interest sooner, which increases mortgage payments in the short term. The lower cost only comes 15 years down the road when the mortgage is paid off sooner.

And anyone with an existing fixed rate mortgage is paying no less, while anyone with an existing adjustable rate mortgage is paying more. They both also lose any equity they had and could otherwise have been able to borrow against for consumption, investment or emergency use.

> Lowering of prices would also allow other investors to take some risks. For example, if the cost of input goods into a factory start reducing, that would free up capital to hire more people or do more R&D.

Except the price of the output goods starts reducing at the same time, so there is no extra money to invest, a hard conversation to be had with employees who will have to have their salaries proportionally reduced (even though they still have to make the same mortgage payment), and a serious problem if you have any inputs whose price hasn't declined, e.g. due to long-term fixed-price contracts or existing inventory that was bought on credit.

> Capital freed up because of lower asset prices will circulate in the economy again.

It isn't really a matter of being freed up. At a given money supply, that's how much money people have in total. If the price of a transaction is lowered then the buyer has more and the seller has less, but the total is still the same.

The issue with deflation is that it starts with the net destruction of money. So the buyer has less because it was destroyed, then the seller gets less because the buyer can't afford as much. They both have less. Then the next seller in the chain also gets less. All the prices have to be lowered for everything, but that takes time to shake out and in the meantime people can't afford rent, default on their obligations, etc. -- which destroys more money and triggers another cycle.

The theory that there is a floor to it because in the limit people will pay whatever they have to for necessities doesn't even work, because if you're unemployed with an underwater mortgage you can't afford, that happens on day one and on day two the amount you have left to pay for necessities is zero. The limit is hunger riots and anarchy. It's the Great Depression. Getting within a hundred miles of the limit is doom.

> At the essence of this discussion, we are picking winners and losers. I'm not a fan of government making those picks.

Neither am I, but unless you're going to abolish the central bank, it has to set interest rates at some level and release some amount of money into circulation in some way and those choices will affect things. "If you choose not to decide you still have made a choice."

> The lenders need to take some risk and only that way will speculation end.

They did what we induced them to do. The original housing crisis was caused in large part by government policies promoting subprime mortgage lending, to allow lower income people to afford a home. The current housing bubble was caused by a government policy of low interest rates to shore up housing prices and save people from underwater mortgages. What sense does it make to purposely induce the banks to do something and then punish them for it?

> For example, without bitcoin bust this year, people would've kept speculating the value of bitcoin to the moon (as was happening last year). With this bust, bitcoin denominated assets are still denominated the same but on real terms, they are deflated.

No, no, it's just the opposite. Bitcoin is down -- it's the currency -- so asset prices in Bitcoin are up. You now have to pay more Bitcoin for the same ham sandwich. Bitcoin prices rising to the sky is the deflation spiral, prices crashing is hyperinflation. The same as the dollar loses value against the Euro (or vice versa) if there is more inflation in one region than the other.

It's one of the other problems with deflation, and why Bitcoin is designed wrong. It's intrinsically deflationary -- the currency supply can't increase in proportion to demand for currency, so you get deflation as currency demand rises. But deflation causes currency speculation, so you get hyper-deflation, i.e. Bitcoin prices soar. As soon as the demand starts to fall off at that price, so does the speculation, so the currency crashes. Then people start talking about "the death of Bitcoin" even though its utility for non-speculators is the same as it ever was.

If some actually useful applications of blockchain start to appear before Bitcoin is displaced by some non-deflationary alternative there will be another rise and fall. Eventually people will tire of that and fix/replace it with something that allows currency supply to respond proportionally to demand.

> To conclude, while defaults sound bad, they're not as bad after such a bubble is created (so long as they don't cause systemic faults, at which point, the system should switch to minor inflation, UBI)

I would agree with that -- something small enough not to have systemic effects should be allowed to fail as nature intended. But consumer debt is so large that it's inherently systemic. And it's a creation of government policy. Uncle Sam needs to clean up the mess it created.


GE could fail. They are on the way down.


Each recession is different, why do you think AI would be able to predict them? Where would it learn to predict from? When using AI on a time series, it is necessary to have sufficient past data.


Matt Levine has a half-joking article on whether or not automated trading would one day provide the ultimate fulfillment of Marxism:

https://www.bloomberg.com/opinion/articles/2016-08-24/are-in...

If you believe that AI will get better at allocating capital over time, and that central planners are historically awful because of human flaws (from self-interest to lack of information)... then you might one day reasonably pine for self-driving economies.

Asimov slipped this general idea in some of his stories as well:

https://en.wikipedia.org/wiki/The_Evitable_Conflict

Even if you think this idea is completely silly (there's an argument it is), the Levine article is still highly recommended, primarily because he's not evangelizing so much as just toying with the idea. It's a fun read.


This kind of wishful thinking by people who dont realize there will always a human controlling that AI.


I think maybe that is a wishful idea from a human :)


It is already being applied to finance. The problem is that a huge part of the market is based on emotion, so the AI makes it worse and not better.

When the proverbial knife is dropping, you don't want to catch it while it is in free fall, it will likely plunge below "fair value" in a panic.

Quants know this and most have programmed/taught their systems to pull out of the market or short when volatility goes up. It is happening right now. Liquidity since February has shriveled up.


AI tends to "over-optimize" as we know from the paper clip theory too. It's likely to make booms and busts even more drastic.


A recession is coming in the next 2-3 years due to demographics (peak baby boomer retiree is happening right around now). I think part of the housing boom is caused by the retirees.


A recession is due for lots of reasons, like the boom has continued this long because we juiced the economy with a bunch of tax cuts for rich people and companies that are basically a sugar high. At least since we raised interest rates we'll be able to cut them to stimulate the economy. On top of that sugar high we are pushing it down with stupid tariffs on most of our major allies. And growth periods in our history don't usually run this far. I think our future growth in the us will be significantly retarded because of high student loan debt. That gets some attention here at hacker news, but I don't see 'regular people' talking about it much.


How? Retirees aren't buying houses, they're selling them.


Time to get out while you still can.

A lot of the current value is tied up in stuff that will implode like in 2006.

The Fed has created another asset bubble. We are sitting on a powderkeg.

But where to invest instead?


> But where to invest instead?

There's literally nowhere you won't lose money when investing in a recession, the point is to find the place where you'll lose the least amount of money -- a.k.a. the place where you are most certain the price has bottomed out -- and go for it. Then wait 5-10 years.


What about a basket of really promising startups or VCs?


I presume this is a joke, but, if not, good luck. If you can pick the next Google reliably, you're presumably already worth tens of millions.


How about a vanguard index of stoks


You need a mixture of assets. If you don't need your investments to live on (cause you've got a job and an affordable place to live), then follow the standard policy of different categories of stocks, plus other areas.

In 2007-2009 I noticed that I lost as much in my retirement accounts as I put into them in those years, Ie the value at the end of the year after putting 20k was about the same as at the start. That was something that really reminded me that it was great and lucky to have a good job as a software engineer.


The trick is to maintain your job during the recessions and keep investing. Easier said than done, but if you do succeed, you'll be in very good shape when the market bounces back during the next cycle.


That's very difficult in normal times (vcs mostly pick losers, just get a few winners and they are doing good), so for regular people we can't handle that much loss or get good returns on winners. When businesses stop expanding (cause there's a recession!) it's even more impossible for startups to succeed.


Maybe it is not an anomaly, but a message: the huge amount of liquidity pumped into the system is going to break soon in the form of hyperinflation. [0]

[0] https://www.google.de/amp/s/seekingalpha.com/amp/article/404...


The Fed is planning to shrink its balance sheet by $2 Trillion between now and 2022. They've already shrunk it by almost $350 billion (basically, shredding digital cash.) It is more likely the global economy is more fragile than the Fed realizes (due to China) and we will experience asset deflation, not inflation. We likely (imo) won't see any significant velocity of money changes in the US until well after the unwind is complete. Because banks, while flush with cash, are still hesitant to lend during the unwind and high yield credit markets are not doing well at all.

I suspect we could see a mild recession with little Fed action other than just pausing their unwind for a while, unless there is some serious crisis (which I can't see happening unless the high yield credit market were to completely implode.) One problem that could allow that to happen, is if oil were to completely and utterly nosedive and make a lot of oil service companies with high debt loads default. Again, I doubt this. I am not a doom and gloomer, I think we are about to enter a recession/bear market but the "great recession" has gotten something in people's heads that every recession must be a systemic collapse instead of a resource reallocation...


If we have deflation with rising incomes, there would be no issue. If there is deflation with declining incomes, it would create a downward spiral, potentially.


Real incomes have fallen during/after virtually every US recession; at least in -real- terms. Nominal incomes almost never fall and just remain flat. Monetary theory is complex, but what the flow of credit remains in tact.

https://fred.stlouisfed.org/series/MEHOINUSA672N

https://fred.stlouisfed.org/series/MEHOINUSA646N

The only way I can see hyperinflation in the US is some cataclysm or inability to service debt through normal means. Most countries that have hyperinflation generally have massive debt loads they are printing out of with no economic output to back it up. They also are usually very centrally planned economies or war torn.


When employers contract their payroll, they do so with layoffs rather than pay cuts. The reasons for this are fairly complex, but if we take this as a given it implies a lot about our monetary policy. Specifically, it argues for something like NGDP targeting, so that employers as a whole continue to have enough money to avoid layoffs even as total real wages decline.


Not necessarily, the whole reason we invaded Iraq was to maintain the petrodollar.

The Saudis would be richer if they unpegged the Riyal and exclusively sold their oil in their currency.

The world is essentially the US economy.


Bonds also mature.

The Federal Reserve’s balance sheet primarily contains bonds. They just get the principle back and the balance sheet is reduced. They dont have to sell on the open market.


That article (from just shy of two years ago) seems flawed. The premise seems to be that higher interest rates would increase the velocity of money, which would cause a lot of inflation given the current (large) money supply and low velocity. But higher interest rates reduce inflation, because their dominant effect is to cause people to borrow (i.e. create) less money, or pay back (i.e. destroy) money they've already borrowed, as soon as the interest rate exceeds the time value of the thing they borrowed the money to do.

That does increase the average velocity of money, because the money that gets not-borrowed or destroyed is the money with the least productivity, i.e. the money that was just sitting around doing nothing. The lowest velocity money. So when that money is destroyed, the average velocity of the remaining money is higher. But the average velocity of the remaining money hasn't actually changed, you've just destroyed the lowest velocity money. V times M is lower, not higher, because the dominant consequence is less M, not more V.


It would be nice to see an analysis of the cost to build over time. Obviously, if 5% YoY increase is the new method of valuation accepted by all (per the article) there must come a time when builders create new units and compete on price.


Not when the majority of the price is in the value of land.


As a 30 yo male who doesn’t intend to have kids, I’m wondering if it is a bad idea to buy a 1 bedroom in sf even though the math doesn’t seem to work out right now and 1 bedrooms historically do not appreciate as fast as multi bedroom homes.


I think this depends on your math. If you are willing to continually move to cheaper rents and invest your money to get better than X% year-over-year, then renting is a great option.

The hidden value is in stability in having a home, which is very subjective.


Owning also helps close your natural short on housing. Your real cost of living gets eaten by inflation over time.


It's probably fine. Does the implied rent cover the mortgage?


buy a multi unit, you can have 1 or 2 rooms and rent out the rest.


That sounds like a job


Housing Booms is the United States longest running con. It makes no sense, and is so completely disconnected from reality it creates a separate one most American's participate: the housing lottery for retirement.


It's funny cause I just bought my house in 2016 and now prices are declining.. Can't win.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: