I know that the “low end” of the income distribution isn’t buying homes.

I also know that the upper tiers are not as constricted to the 3x income thing…however…those top earners are not only buying just one home. Someone has to buy all those ski and beach condos…not to mention, the rental units for the low income earners.

Per deejayoh’s chart @ 15, instead of 5.25, the baseline becomes around 4.0, so if you assume some possible overshoot, the bottom would be between $224k (3.25x) and $275k (4.0x).

ahhh, but in the late 90s you could have said the same to someone who said home prices were going to 8X incomes!

“My main point is that the median house value to the median income ratio is always going to be far removed from the criterion used to evaluate first time homebuyers.”

This I kind of agree with but the median price reported is for homes sold, not homes that are not. I don’t think it’s median income / median value I think it is median income / median price that is tracked?

The same kind of goes for this comment of yours:
“Don’t expect that group to dump their houses in fear that prices are heading towards 3 times income. ”

If these homes are not part of the “market” as in homes that trade hands they have no impact on the market. But when these homes do change hands they need to follow the current market prices independent of if they are paid off or not.

Not sure why you are using per capita income, the comparisons I have seen in other analysis that refer to the 3X ratio are Median Family income to Median household price. Someone earlier mentioned median family income in kind county was 70k, which I think is a little low based on prior research, which I didn’t keep, seems like it was 76k, which would give you a house price of around $230k, which would be comparable to your estimate, of 5 X per capita income. Median family income is likely more accurate as it accounts for dual family incomes.

Canary – I agree that HHI is a better measure of income. I used PCI because a) the trend tracked HHI very closely and b) I had 20 years more data! I was interested in showing the strength of the underlying tie between income and home prices – so the longer series worked better for me. The longest HHI series I could find was back to 1989. BTW – as of 2008, HHI for KingCo was $68k. Check the OFM site.

But you can’t carry the multiples from this analysis over and apply them to HHI. The results of the comparison are different – and the long-run multiple is more like 4x with a low of 3.5x. I showed a chart comparing HHI to median in comment 15 if you are interested. So if you use that data, the expected “bottom” for housing would be between $238 and $272k – which is pretty close to what I get using PCI. That’s why I was indifferent to using PCI

Per Capita is the Average, in this case of income, not the median so it will matter.

http://en.wikipedia.org/wiki/Per_capita
http://en.wikipedia.org/wiki/Per_capita_income

“Per capita income gives no indication of the distribution of that income within the country, so a small wealthy class can increase the measured per-capita income far above that of the majority of the population.”

From the PDF I posted above we know this is the actuality over the period in question in the analysis.

Willie says:
If one were to assume that the income distribution of the US as a whole is significantly representative of King county, then backing out these top 1% and their respective income percentages from the PCI would give a much more representative picture of the other 99% of the population.

If one were to do so it appears it would show prices more out of the ordinary currently and subsequently with a more significant correction in order.

This is false.

The whole reason why you use median numbers instead of averages is to minimize the effect of top/bottom 1% inluencing the numbers. It does not matter if the top 1% gained 1000X growth in income instead of 10X, unless their numbers increased, it would have no effect on median.

I also believe too many here are focusing on minor details. The fact that the chart show is that there is a VERY VERY strong correlation between personal income and housing prices. It does not matter if you think HHI would be better or something else would be a better indicator. The numbers show that PCI is excellent as is.

Don’t make it more complicated than it needs to be.

Interesting chart. People do seem to think that those years of high appreciation are a normal occurance. Even here on the Bubble people try to use those years between 2002 and 2007 as a part of a larger equation.

Housing starts and construction spending became such a huge part of the economy during that time. Job loss is the perfect word because we will never again see the amount of construction we had during those years.

For quite some time, I’ve had my eye on Seattle-area house prices declining to about year 2000 levels. I don’t have data to back up my suspicions. It’s either a gut feeling or purely wishful thinking. House prices seemed affordable to me back then, and they don’t now. Even with the latest discounts, current prices are crazy high. I agree with you though that what ultimately matters is where median incomes are headed, and that we’ll eventually adjust toward them.

Interestingly, when a lot of people I know speak about “getting back to a normal economy,” they are referring to getting back to a completely abnormal period of bubble-mania. This chart says a lot about what it will take to return back to where we were a few years ago:

http://graphics8.nytimes.com/images/2009/05/08/business/economy/joblosses.jpg

Who knows who is right. Only time will tell

Duh, I get it. $100K is meant as the prediction of median price at bottom–not the predicted amount of drop from current.

Hmmm, looks like I’m not a super-bear. Then again, I’m not exactly an optimist either.

Great analysis!

“The April median for King County SFH was $380k, indicating a possibility of 22-42% of additional downside risk ”

I believe the historical homeownershiip rate will factor into the ultimate bottom. IOW, if we are at the upper end up your range, and the homeownership rate is still well above normal, I expect prices to keep falling.

If I’m reading you correctly, a 22-42% drop represents a potential for a further $83K-$151K off the median price.

“For home prices to fall much further than this (e.g. the super-bear’s $100k prediction), it seems to me that one or more of the following things would have to be true”

OK, it looks like I’ll have to read it again.

This PDF: http://www.investmentpostcards.com/wp-content/uploads/2009/05/green-shoots-or-smoking-weed.pdf
shows in chart 1 that according to the IRS from 1969 to (I think) 2006 the top 1% of income earners saw their percentage of national income rise from approx 10% to 23% respectively.

If one were to assume that the income distribution of the US as a whole is significantly representative of King county, then backing out these top 1% and their respective income percentages from the PCI would give a much more representative picture of the other 99% of the population.

If one were to do so it appears it would show prices more out of the ordinary currently and subsequently with a more significant correction in order.

Check out: http://www.economics.harvard.edu/files/faculty/51_Aftermath.pdf

After a financial crisis, RE prices fall, on average, 35% and the duration, peak to trough is 6 years. The worst scenario is 50% drop lasting 10 years.

Looking at the available trends, I believe that housing prices of the 1997 – 1998 period are a good guide for the bottom of the market in general and for specific locations. I wouldn’t pay more.

I would be more concerned about those underwater on their mortgages, particularly the subset who took out several HELOCs, and the other subsets who may be suffering loss of income and/or resetting ARMs. I wouldn’t even expect these to sell in fear of price drops, but rather in fear of foreclosure and the social stigma of not being as rich as they were before (or pretended to be).

I don’t see why home prices in the Puget Sound won’t return to at least the late ’90s levels.

Well, he did say it wasn’t meant to be predictive, but Deejayoh’s stated potential “bottom” range of $220k to $295k corresponds with the (non-inflation-adjusted) medians of July 1998 and June 2003, respectively. So based on this analysis, late ’90s pricing doesn’t actually seem entirely out of the question.

Don’t expect that group to dump their houses in fear that prices are heading towards 3 times income.

2004 Median household income of owners: 72,009, renters 34,268.

Page 36 breaks down mortgage vs. no mortgage.

I think that document has all the other info you ask for as well, as of 2004.

Do you know if there is data available related to:
-average income for homeowners
-average income for non-homeowners
-number of owners vs. renters or as a % of total in King County
-% of homes that are owned outright
-% of homes that are owned by retirees
-average time since purchase by homeowners in King County

I have a feeling that the numbers are skewed by the mix of these variables. ie. retirees with low incomes who own a house they bought in Magnolia in 1969 which is now worth $850,000, or the median income being skewed by some very low data points (these people never bought in the past and never plan to… nor should they), ect ect.

While 1969-2009 saw some economic turbulence, it was nothing compared to what happened during 1920-1960 . I don’t think there’s any doubt the next decade will more closely resemble the 1920-1960 period.

I tend to agree with Scotsman. I think the last 40 years may not be a good comparison for what is lying in store. When you look at home appreciation for the last 100 years, the post WWII years stand out like a sore thumb. The developing world has experienced over 2 generations of almost continuous economic prosperity, which is (historically speaking) an extrraordinary aberration.

Here is a look at the average US home price increase by decade for the last 100 years.

1890’s 0.53%
1900’s 1.40%
1910’s 3.30%
1920’s -0.70%
1930’s -0.45%
1940’s 8.16%
1950’s 2.67%
1960’s 2.57%
1970’s 8.12%
1980’s 5.86%
1990’s 2.84%
2000+ 9.27%

I don’t see why home prices in the Puget Sound won’t return to at least the late ’90s levels. Wages have barely gained since 1999, and are now in a tailspin (just ask anyone looking for a job right now how much lower the salaries compared to their previous employment). It is also clear that the massive sprike in prices since 2000 was unprecedented even by the unprecedented appreciation post WWII appreciation rates. I see no reason we can’t lose all the gains of the last decade.

Of course, I think prices will go even lower than 1999 levels, but that’s another topic of discussion…

I’m amazed this year that $250K to $350K buys you a two bedroom, one bath house, and people think that is great.

The price of property has nothing to do with your income or interest rates. It has to do with the value of the property. We have a long ways to go to address value.

Town houses should be cheap housing units. That’s what they were built for, that’s what we permitted them for. When they sell for $300K that compromises all property pricing. It drags in an expensive price per square foot.

The money you spend should buy you something.

The first time I saw a Target Store I made the comment that it is the most expensive store I had ever been in. You pay a very high price for really cheap carp. There again Nordstroms has followed that business model today so there must be something to it.

I dunno, I think 3x HHI sounds pretty reasonable even if in our case i would regard it as a stretch. It would mean a mid $600k home which would require a very,very low interest rate to make it comfortable for us if we want to continue paying for daycare, 401ks and college funds and keep a positive savings rate not counting the 401k and college funds. Not to mention the down payment required,we happen to have it but I think it’s pretty rare. 3x HHI actually seems a bit high for families. Dinks should be ok so perhaps it averages out. But 5x would take us over $1m which I regards as impossible and completely financial suicide.

And please add health care costs.

By they way, Beaver’s dad not only paid 2-3x income; but he had a pension (no need for a 401K), incredible job stability, and an amazing health plan. Those were the days; if the world needed from wire hangers to lollipops, they had to buy them from the US. Boy, are those days over….in fact back then the biggest employer in the US was…..GM. Few people realize their bankruptcy marks not only the end of a company; but the end of an era for us.

I present the following arguments:
a) PCI is averaged across owners and renters. Owners can afford to buy because they earn more. If you could get PCI for the top 70% of earners, then the PCI would be higher and the Price:PCI ratio lower.

b) PCI does not take into account dual income households. That will also skew the Price:PCI higher. Again, if you could calculate the household PCI, your income distributions would be higher and the ration lower.

I suspect the potential botton is closer to 3x income.

The old established “rule” is for mortgage payment to be at or below 28% of one’s gross income (not purchase price).

The top 1-2 percent of earners can now afford much more expensive homes (correcting for inflation) than their counterparts could in 1969. But wages for the rest (bottom 90 percent) have not kept pace.

Second question: Are the multiples likely to stay somewhat higher now partly due to the growing income gap in Seattle and King County since 1969? Or is that a silly question?

You take the mortgage payment based on a thirty year fixed mortgage or fifteen year and you get the price of the property.

Tax and insurance added in gives you a mortgage payment of about $1100 for a $50K Income, $2200 for $100K.

How are we doing with that?

http://www.economicpopulist.org/?q=content/subprime-meltdown-over-now-comes-bad-news

Maybe that means I’ll never be a homeowner – I’ll always be outbid by those willing to take bigger risks. In the new economy if home prices stabilize at 5-6x income, ownership seems like a big gamble instead of a road to financial security.

I guess this is where the lower income population that rents acts as a buffer and make the higher ratios statistically possible.

No one has ever claimed that home prices should normally be 3x per capita income. As has been pointed out above, it is 3x household income, or family income.

well, actually…

http://seattlebubble.com/blog/2009/04/02/seattle-homes-still-10-20-overpriced-compared-to-rents-and-incomes/#comment-69885
http://seattlebubble.com/blog/2008/07/02/whats-your-housing-bust-strategy/#comment-51078
http://seattlebubble.com/blog/2008/07/02/whats-your-housing-bust-strategy/#comment-51016

I agree with the point that the analysis should be on HHI and not PCI, but I can say with with some certainty the median home price has not been 3x median or average HHI at any time in a period of history that is relevant to today. Maybe if you are old enough to have watched the premier of “Leave it to Beaver” it was

I too have watched the incomes [household] mysteriously rise with high unemployment/underemployment and wondered how this could happen…..we know that a lot of kids/parents are moving together with kids/parents in this recession; could it be when that happens you have 1.4 single incomes per average household, rather than the old 1.2 single incomes per average household?

This means we could actually be making less per capita; but household incomes rise from like $40K average to the present $50K?

Since there are twice as many single households with just single incomes; perhaps you could make a revised chart for the mainstream household: divorced and never married. We rule you know…LOL….

I’m sure it makes the comglomorate optimistic married charts you used look way “overly optimistic” for the majority of single income households in Seattle….we’re probably the 1st time homebuyers only hope too, as even two young adult incomes with young kids and daycare expenses really have about one income anyway….

No one has ever claimed that home prices should normally be 3x per capita income. As has been pointed out above, it is 3x household income, or family income.

Per capita income is really not the statistic which should be used, especially if you’re looking over a long period of time, since changes in family size or changes in the distribution of wealth in society (people on the top becoming billionaires) can throw this average way off.

It’s difficult finding the stats you need to properly do this sort of thing, though, I understand.

Still, when I look at that graph I see a steady accumulation of wealth, apart from a few deviations. For example from 1991 to present the housing stock has gained 2 years income worth of value over an 18 year time span. For a long term asset class like real estate, that is not too alarming a number. If houses were discarded at the end of one year, then you could be more concerned about it.

As a member of a cohort near the peak of the baby boom I look at these things as being driven also by where I and my peers are in the lifecycle of housing needs because that group often shows up in the overall statistics. During that time we bought houses and endeavored to move into bigger houses. Once they were all in their main family houses, the cycle of rising prices ended and the bubble collapsed. But the younger cohorts are nearly the same size as the peak baby boom. It is the older cohorts that are smaller. So that is why the graph shifted and will not necessarily shift back.

http://www.bea.gov/newsreleases/national/pi/pinewsrelease.htm

I don’t see where this is available at the county level. I think that is an annual series only.

I believe it was your mortgage should be no more than 3x your income. Move up buyers and those with larger downpayments will obviously be more than 3x income.

This morning there was another report on how “special” this recession was because of the large number of people with degrees out of work for very extended periods.

Back to old school; if you are part of a 2 income household, make sure you can make ends meet with one salary.