I got a lot of pushback when I wrote last month that San Diego home prices were, on the whole, back in line with their historical relationships with local incomes and rents. One of the more frequent counterpoints was the claim that while home prices might be in line with rents, rents themselves had become unsustainably high.
How to analyze such a question? My first thought was to compare how much rents had changed in comparison to local incomes. An increase in rents that was way out of line with what San Diegans were earning would suggest the rent prices had indeed become distorted by the housing bubble.
But the data I have indicates that this is not the case. The accompanying graph charts average San Diego rent as a percent of income per person. The average percentage over the entire measurement period is illustrated by the orange line.
Very good info Rich. I
Very good info Rich. I wouldn’t have guessed that we’re actually below historical average. I have a few questions. Does unemployment affect per capita income number(i.e. do they count as $0/yr. or not get counted at all)? If Obama follows through with this promise of raising minimum wage, wouldn’t that make a big impact in per capita income? Since, I’d assume that, those who are currently making a little above minimum wage would also get a pay bump to stay slightly above the new minimum wage as well.
Is this gross or net per
Is this gross or net per capita income?
Good to see real data. I am
Good to see real data. I am surprised too. When I lived in OC, my rent doubled from 1996 to 2008. I didn’t think incomes had doubled. Maybe a 50% increase.
PR – Here’s the changes in my
PR – Here’s the changes in my SD data from 96-08:
Rents – up 87% (so this matches your experience pretty well)
Per capita income – up 83%
FWIW…
I will post a graph of the separate income and rent series in case that’s interesting for anyone.
Rich
rents and incomes
[img_assist|nid=10910|title=rents and incomes|desc=|link=node|align=left|width=452|height=329]
Wow, that’s educational. Per
Wow, that’s educational. Per capita incomes up 83% from 1996. I’ll accept that as is, but I am open-minded about what drove it up so high. I wonder if it’s something unsustainable like higher workforce participation rates, or illegal immigrants counted in income but not population, or large capital gains in more recent years…
patientrenter wrote:Wow,
[quote=patientrenter]Wow, that’s educational. Per capita incomes up 83% from 1996. I’ll accept that as is, but I am open-minded about what drove it up so high. I wonder if it’s something unsustainable like higher workforce participation rates, or illegal immigrants counted in income but not population, or large capital gains in more recent years…[/quote]
Does the way that we count income by pop actually lend itself to such distortions?
Who has knowledge of the process?
Rich I am hoping?
Maybe I’m not reading this
Maybe I’m not reading this correctly, but it looks as though rents never seem to get below about 42% of income. Yet, most DTI ratio’s for mortgages are in the low 40% level at about their highest. If this has been consistant over time, then the average income has never been able to get an average loan? Or perhaps the average mortgage has been less than the average rent, over the long haul?
peterb wrote:Maybe I’m not
[quote=peterb]Maybe I’m not reading this correctly, but it looks as though rents never seem to get below about 42% of income. [/quote]
That is 42% of _per capita_ income. If a typical rental unit is occupied, on average, by two people, that would mean that rents average 25% of household income.
Though on the other hand, renters usually have lower incomes than general population.
[quote]. I wonder if it’s something unsustainable like higher workforce participation rates, or illegal immigrants counted in income but not population, or large capital gains in more recent years…[/quote]
It’s not higher workforce participation rates. Those have been fairly stable since Reagan era.
Since this is per capita income, it may be disproportionately affected by high-end growth. One person’s income growth from 120K to 130K (8%) matters as much as someone else’s growth from 30K to 40K (33%). We did have a lot of high-end growth in 2000-2008, in biotech, telecom, to some extent in software. There wasn’t much high-tech employment here in 1996, as far as I know.
The same is true for average rent.
Also using 1996 as a starting point would be a bit misleading, that was the bottom of the last recession and incomes were depressed by above-average unemployment.
Median household income grew 45% between 2000 and 2008, which is barely above inflation.
Great Discussion.
In
Great Discussion.
In summary, what I am getting is that while Rich’s graph suggests rents are as “fair” now as they have always been (since the 70’s), it does not consider that renters’ incomes have not gone up as fast as the per capita income – because the per capita average is skewed up by the relatively few big upper tier incomes that went up a lot. And it does not consider that renter household size has perhaps increased to keep pace…
Teasing out these factors could lead to even more valuable analysis of current rent “cost to renter”
Eugene wrote:
It’s not higher
[quote=Eugene]
It’s not higher workforce participation rates. Those have been fairly stable since Reagan era.
Median household income grew 45% between 2000 and 2008, which is barely above inflation.[/quote]
Good points Eugene. Just want to add the LPF has changed in last 9 years, just a point, but that can skew unemployment rate, etc.
Civilian Labor force 16+ was 142,434,000 Mar 2000, and March 2009 is now at 154,048,000.
[img_assist|nid=10920|title=96-2008 US Labor Participation Rate|desc=|link=node|align=left|width=466|height=233]
http://data.bls.gov/PDQ/servlet/SurveyOutputServlet?data_tool=latest_numbers&series_id=LNS11300000
I think what needs to be
I think what needs to be considered is the price of a place and what it would rent for at this time. All this conjecture about average per-capita rent and income is kind of “noise”. Too many hidden variables. Especially when stats from the BLS and other govt agencies are used. That’s why the CS index is so much more meaningful to the subject of this site. Compare apples to apples.
patientrenter wrote:Wow,
[quote=patientrenter]Wow, that’s educational. Per capita incomes up 83% from 1996. I’ll accept that as is, but I am open-minded about what drove it up so high. I wonder if it’s something unsustainable like higher workforce participation rates, or illegal immigrants counted in income but not population, or large capital gains in more recent years…[/quote]
I’m confident in Rich’s ability to read the data, but that data doesn’t quite pass the sniff test. People doubled their income from 1996 to 2006?
Maybe it does pass the sniff test. People lived high, brokers, RE agents, you name it were totally making bank. People had massive stock market gains. There’s a reason county governments are all projecting double digit drops in sales tax revenue…
Sorry Rich, I don’t think I can help on the income data, the obvious source doesn’t go back that far with easy access.
Oh, one more thought. Is that San Diego, San Diego city or San Diego county?
My income-income did not
My income-income did not double, not by any stretch of the imagination. Take away any income from investment sources. Income from a working job. Mine has increased slightly, but not doubled. However, car, rent, gas, etc, all went up measurably, some doubled.
NSR: It’s the San
NSR: It’s the San Diego/Carlsbad/San Marcos MSA.
PR: To your list of potential causes I would also add the rise of telecom, biotech, etc… some high paying industries have flourished here. This fact was abused by the bulls in trying to justify infinitely high home prices… but it is legitimately true that a lot of high income jobs have been created.
UR: Sorry, don’t have any particular insight.
Peter: Household income would be the correct stat to measure against mortgages if you are trying to determine affordability. This is just to show where things stand compared to history.
sobmaz: I use the CPI rent component which is not affected, for the most part, by the issues Williams has identified with the CPI. (As a side note, while I am sympathetic to his goal I find his alternate CPI to be even less realistic than the BLS one).
JP: This doesn’t necessarily mean that everyone’s income should have doubled… as I mentioned above it could be that there was a compositional shift towards higher paying jobs.
Generally: If anyone wants to play around with the data it is found here: http://www.bea.gov/regional/REMDchart/default.cfm#chart_top
Rich
Thanks Rich, it’s one of the
Thanks Rich, it’s one of the sniggly details that trips me up when looking at the data. Any chance to utilize the people per household stats modify the income numbers as a stand-in?
Can you utilize the division of the owner versus renter stats from the census to differeniate people per household numbers? Then post compute back to a renter household income by per capita to see how population density has changed specific to the renter pool?
no_such_reality
[quote=no_such_reality][quote=patientrenter]Wow, that’s educational. Per capita incomes up 83% from 1996. I’ll accept that as is, but I am open-minded about what drove it up so high. I wonder if it’s something unsustainable like higher workforce participation rates, or illegal immigrants counted in income but not population, or large capital gains in more recent years…[/quote]
I’m confident in Rich’s ability to read the data, but that data doesn’t quite pass the sniff test. People doubled their income from 1996 to 2006?
Maybe it does pass the sniff test. People lived high, brokers, RE agents, you name it were totally making bank. People had massive stock market gains. There’s a reason county governments are all projecting double digit drops in sales tax revenue…
Sorry Rich, I don’t think I can help on the income data, the obvious source doesn’t go back that far with easy access.
Oh, one more thought. Is that San Diego, San Diego city or San Diego county?
[/quote]
I pulled out our Annual Social Security statements. I was a bit surprised in the discrepancy in my memory versus the actual hard numbers. (I urge others to do the same). I was expecting about a doubling of income based on memory. Our income tripled from 1996-2008, but we were fresh out of college and on the steep part of the income curve as young professionals. (not working in real estate) The curve certainly has flattened for us in the past couple years, though.
But, I think that the fact that income growth has been skewed towards the higher end over the past decade might be an important reason why rents are below long term averages and might be the root cause of the disbelief in the per capita income growth measured expressed on this board.
Regardless, if you somehow adjusted for these factors, I seriously doubt that rents are higher than long-term averages in San Diego when compared the past 30-40 years.
Looking forward I would expect income levels to flatten and rents to fall as the current economy is accounted and the 2009 numbers are released (charts are through 2008).
Also, there is certainly room for a 10% drop in relative rents from current levels if we were to reach the same point as we did in the late 1990s.
I began renting our apartment
I began renting our apartment in August for 1200 per month. 1br 1ba. I saw our complex advertising for 900 per month just yesterday.
A better comparison would be
A better comparison would be household income for rent.
Per capita gets distorted by family size versus individual income.
Rent/household will give you the percentage of income for the household going to rent.
Rent rates above 30% are poverty inducing.
If delflation hits wages,
If delflation hits wages, then the bar will have to be adjusted downward. Unemployment is a type of wage deflation when considered in the aggregate. Using $ numbers would be more meaningful as there’s a lot of “stuff” hidden in percentages. Especially the one’s provided by the BLS.
AN (and Peter): Per capita
AN (and Peter): Per capita income is all the income generated in the region divided by the number of people. So if unemployment goes up, all things equal that would reduce per capita income.
Alice: Gross, if I understand the question.
NSR: I agree but I do not have good sources for historical household income… do you know where i could find it for SD?
Peter: Same answer (on your actual $ question) — the CPI rent component is the only data series I have found that tracks historical rent prices. If you know of another source I will definitely check it out.
Rich
One other point I meant to
One other point I meant to make on the household vs. per capita income question: over time, household income has risen in comparison to per capita income as I’m certain there are more 2-income households now than in the 70s. So if we used household income, the ratio would have fallen even more from the beginning to the end of the graph.
Rich
Rich:
Long time reader and
Rich:
Long time reader and haven’t commented in a long time. I am a native San Diegan, Point Loma specifically, over 50 years old and can remember when San Diego housing was less expensive than that of the Chicago area (many relatives of mine live there). This was due to no basements and exterior brick work, hence lower actual construction costs. The sunshine tax, to my recollection, has maybe only existed since the early 70’s when real estate began to move up in earnest here and surpassed the midwest in price. I would want to see data going back to the early 1900’s to make any claims of normalcy with regards to home prices/payments to incomes and rents. The 1970’s was a very inflationary period where assets such as housing in SD eclipsed the inflation rate. I believe that has distorted the true “historic” nature of these ratios. In other words, the data from the mid to late 70’s doesn’t go back far enough to gain a true picture.
Peterh – I don’t have data
Peterh – I don’t have data going back any farther.
I’m not sure I agree with the premise anyway… industries/popularity/etc of cities changes a lot over time, so the further back in time you go, the less relevant such a ratio would seem as far as what it should be today.
We are looking at what comprises a “sustainable” ratio, and if the ratio has been similar for decades that seems sustainable enough — even if it had been lower before that.
Rich
If you used the CPI in your
If you used the CPI in your calculations did you use the FAKE CPI that the government now uses or did you use the REAL CPI data prior to Clinton?
Their massaging of the numbers has made the new or FAKE CPI about 50% below what the old or REAL CPI use to be.
You can get the specifics at http://www.shadowstats.com/
Most all stats have been changed over the years, from unemployment to GDP, so when a person attempts to measure a constant using floating perameters is is rather difficult and sometimes highly inaccurate. I don’t know all the inputs you used but just a comment.
I don’t know of anyone who
I don’t know of anyone who has increased their wages even close to that stat.
Something is skewing things. Besides an average is an average an most people make well below average while a few make way above average.
Now those who make way above average have definitely seen their incomes skyrocket the last ten years but certainly not the average working man who pays rent.
Lets just take a look at the minimum wage for example. From 1998 to now it is up a whopping 38%. http://74.125.95.132/search?q=cache:0B2TWPyttPsJ:www.dir.ca.gov/Iwc/MinimumWageHistory.htm+minimum+wage+california&cd=1&hl=en&ct=clnk&gl=us
I think if you surveyed the average hourly worker, which most people are you will find increases more in line with the 38%.
If you survey people like real estate agents, mortgage brokers and such (excluding the last year and a half) you will probably find their wages up more than a 100% and they are screwing the average.
I believe the data is somehow
I believe the data is somehow skewed in the 1998 to 2002 timeframe – the chart shows this period as being one of low rents – during this period, rents were increasing RAPIDLY – trust me, I was buying apartments during this timeframe and usually raising the rents soon after purchase – rents were going up quickly all over San Diego – I talked to many renters during this timeframe and also saw the increasing rents in the MLS listings I was watching (rental property listings usually show rents because that is how rational people buy investment property)
incomes must have risen dramatically during this same timeframe for the ratio of rents-incomes to decline – I assume the income figures include income sources that most renters do not have (stock gains from the dot.com bubble, easy money filling out real estate paperwork (also known as realtors), loan brokers doing refis, etc)
I also wonder how income
I also wonder how income distribution has changed for the region over that time period, and how that would affect affordability. I’ve read a lot about how during the past few decades a lot of the income growth has been in the upper tiers, whereas lower income groups have seen little or no growth in real incomes.
That is why I believe people when they say it feels to them like rents are more expensive than before. The concentration of increases of income at the top (most of whom likely do not rent) have likely distorted the picture and made it seem as affordable as before, when the truth is that those that actually rent are paying more than before. I’m not sure what data you would look at to validate this, but I would be interested to find out.
There are a couple of issues
There are a couple of issues which makes analyzing this difficult. One issue is that San Diego’s average household size has probably shrunk over the years. I’m guessing that people have less kids these days?
I found U.S. census data on the average household size here:
http://www.census.gov/population/socdemo/hh-fam/hh4.xls
This is national data. I couldn’t find any data specific to San Diego. On a national level, the average household size decreased by 11% between 1976 to 2008. If San Diego followed this trend, it could help explain why the per capita income data seems higher than you would think.
The other factor, as mentioned earlier, is that there are probably more 2 income families today than there were in the 70’s. This would definitely increase the per capita income. I’m not sure how this would affect the “average wage” data. I found San Diego’s “average wage” date here:
http://www.bea.gov/regional/reis/default.cfm?selTable=CA34§ion=2
Between 1976 and 2007, the “per capita income” outpaced the “average wage” by 29%. Some of this discrepency could be accounted for by the shrinking household size. I’m not sure where the rest of the difference comes from.
good points, ltokuda
good points, ltokuda
A lot of great points being
A lot of great points being brought up. Many would be addressed by using median household income instead of per capita. I have found some census data back to 01 on this, as well as median rent (which will address Peter’s complaints about the CPI).
I will put up a chart later but here is the preview of the data:
2001 21.6%
2002 21.7%
2003 22.5%
2004 23.0%
2005 22.8%
2006 22.8%
2007 22.7%
According to the census, rents grew fast over that period, but so did incomes. How much of that income growth was fake bubble earnings is the question I guess… but that said, there is not much disparity here.
Rich
Thanks, Rich, for your hard
Thanks, Rich, for your hard work, data, candor and honesty. There are some factors that warrant some consideration. I appreciate all your efforts.
It’s always dangerous to use
It’s always dangerous to use “average” values for data that has a high variance. Such is the case for “wages” or compensation. A “high wage earner” making 700 times the “normal” worker can double the average wage for 700 workers. I’m sure SD has quite a few highly compensated “workers”. The use of averaging for rents is more reasonable because the variance is not terribly great.
My rule of thumb is that when average wage data is used to promote an argument, it is no longer a valid yardstick. Median values are much more reliable and representative of wage data.