September 27, 2006
A Reader Asks: Average Appreciation In San Francisco?
A gentle reminder that if you have a question or query that’s not directly related to the topic at hand, please just drop us a note (email: email@example.com). We’ll do our best to address it in a future post (or even answer it directly). That being said, a reader asks:
Does anyone know the average appreciation for SF 'adjusted for inflation' over the past 20 years? Does it even matter if the appreciation is adjusted for inflation?
We don’t have a canned answer for either question, but we’d be willing to bet another reader might have an answer (and opinion) or two.
And while we’re really not trying to be glib or snarky, we have to ask a question in return: why do you want to know? Or rather, how do you plan on using the answer? As if by instinct, we find ourselves reciting the infamous financial disclosure, “past performance is no guarantee of future results"...
First Published: September 27, 2006 6:49 PM
Comments from "Plugged In" Readers
Historically national home values rise with inflation and wage growth.
Booms occur when these prices rise faster then these two metrics. At some point prices rise beyond the ability (or willingness) of workers to pay.
At the peak is reached prices either fall or stagnate while wages and inflation catch up.
David Lereah said. “Keep in mind that over time, home prices rise at the rate of inflation plus one-to-two percentage points"
Liz Ann Sonders, Chief Investment Strategist, Charles Schwab "For the past century, home prices adjusted for inflation have been relatively stable, with long cyclical swings but no long-term trend."
Posted by: badlydrawnbear at September 27, 2006 7:53 PM
If you want a more techincal explantion the Fedral Reserve Bank of San Francisco has a good analysis of the home price fundementals using the price rent ratio.
You can see on the graph in the article that this ratio mirrors exactly the late 80's peak and mid 90's bottom, returning the ratio to back to its norm.
You can also see the current runup in home prices. This graph clearly shows that rents have remained flat, tracking wages and inflation as always, but home prices have broken away from this economic fundemental. Based on past cycles at some point, and no one knows when, the price rent ration will return to its historic norm with either rents rising, which can only happen if inflation and wages rise, or home prices falling to meet rents.
Posted by: Badlydrawnbear at September 27, 2006 8:15 PM
Here is a graph of Historic OFHEO HPI for Northern California
Posted by: sf94102 at September 27, 2006 8:37 PM
This is probably a pipedream, but let’s please try to keep comments related to answering the question, and not get mired in debating the future. We’ll get to that soon enough...
Posted by: SocketSite at September 27, 2006 9:32 PM
sorry SS wasn't trying to be blogging incorrect. I just heard it and was curious if there was any validity to the 'adjusted for inflation' argument. I didn't realize there was a questions@ss email address. Oh and, you're quick. Not sure I understand it any better but will keeping checking in to see if any of it sinks in.
[Editor's Note: No need to apologize. And as always, thanks for "plugging in."]
Posted by: I have a question at September 27, 2006 9:48 PM
Sorry wasn't trying to perdict the future I was just trying to support my comment that prices track inflation.
[Editor's Note: Great link to the FRBSF and no need to apologize; we’re just trying to be a bit proactive on this one. And once again, thanks for “plugging in.”]
Posted by: badlydrawnbear at September 27, 2006 10:03 PM
Let me go out on a limb here, but I'm guessing that potential buyers & bubblesitters (myself included) would try to use this info to divine what real estate in SF "should be" worth given a return to long-term trends. For example, let's say 2000 marked the last year of what's considered a normal market, and a property sold for $400K. If property historically appreciates at 7% per year, then that same property should be worth $600K today. Right?
Posted by: Dude at September 28, 2006 8:58 AM
I don't know that's why I asked. There's a feed a few down (I think you were in that string as well) which talks about 'adjusted for inflation' prices flat-lining are real loses and how much money one would lose - adjusted for inflation, likes its a magic formula us statistically challenged should understand.
I listened to a talk once that said home prices in the Bay Area, 'adjusted for inflation' have only risen something modest amount per year on average and I wondered if there's any validity to that argument or if anyone had heard the same talk.
Posted by: I have a question at September 28, 2006 9:17 AM
Good grief. The reason we normalize for inflation is to see if the investment is getting more valuable, less valuable, or is staying flat. On "The Wire" on HBO the other night, a new inner city teacher asked his class, "If a rocket is going 1000 miles an hour, how far does it travel in an hour?" Blank, empty stares. Then one kid asks, "it this a trick question?" It was not. And neither is this one. Inflation is the devaluation of your curreny. You buy the stuff you need with your currency. So what you need to know is, how much of your "gain" is you doing better, and how much is the dollar doing worse. Simple stuff, really. Real is all that matters, so keep it real. So, how you doing? Badlydrawnbear has the answer.
Posted by: godzillow at September 28, 2006 9:44 AM
First, you’re going to want historical median annual appreciation, not average. In addition, you’re going to want more than 20 years of data since real estate moves in fairly long cycles. Furthermore you’re going to need historical information about general economic conditions, local employment conditions, taxation changes, demographic trends, housing construction trends (square footage, materials), credit conditions etc. And even then, as the glib or snarky SS comment points out, past performance may tell you nothing about future performance. Second, in regard to your question about real vs. nominal, using real returns to judge any investment is a valid and important method of evaluation. Clearly if you are “saving” money in a savings account earning 3% a year while annual inflation is running at 4%, you’re not really saving anything at all and in fact you would be losing money because of the declining purchasing power of your accumulated dollars. The same holds true for money invested in real estate or any investment whatsoever. One can argue that as long the investment remains in real estate assets and will always do so in the future in the same locale (even if properties are traded), then the real value relative to non-real estate assets and other goods in the economy is irrelevant. However, few people will accept that caveat since job or career requirements, health considerations, family obligations, investment opportunities etc can have one switching locales, adjusting portfolio real estate exposure or utilizing mortgage equity withdrawal for non-housing related expenses. Thus, even with real estate that one owns and lives in, the real return is considered more significant than the nominal return.
Posted by: JS at September 28, 2006 11:29 AM
Got it. So basically I need a multi-variable non-linear regression model that takes into account savings rates, tax law changes, average lifespan, quality of air, density of plywood, and wages. While we're at it, let's throw in an earthquake prediction feature. And even then, everyone agrees I can't predict what will happen to real estate prices in SF.
But I disagree and take a simpler approach: when the cost of living, in US dollars, exceeds what people make, in those same dollars, by a margin that they can't afford, people don't buy. They rent - or leave the city. Eventually, incomes either rise or prices have to fall. Regardless of my tax bracket or health status.
Posted by: Dude at September 28, 2006 11:41 AM
I hear this alot, that looking back only X number of years does not give a clear picture. So for those who want a 'long term' (1890-2006) picture of home prices adjusted for inflation compiled by Yale Economist Robert J Shiller, here ya go.
However, I must warn you, that it is not for the faint of heart or for those who have especially intense opinions about the current housing market.
Also, for those who want to go on the attack after viewing the graph please keep in mind that Robert J Shiller created this graph, not me (or socket site)
Posted by: badlydrawnbear at September 28, 2006 2:10 PM
Link doesn't seem to work....
Posted by: Dude at September 28, 2006 2:45 PM
Actually, the only real way to measure the performance of real estate is to track the price of the same house(s) over a long period of time. Someone did this recently for houses in Amsterdam that are over 300 years old-where they have kept great records, apparently- and found that real estate is not a very good investment over the long term and can have significant short term price swings. In the Amsterdam study (no link, sorry) there were times when the same house could be purchased for less money (inflation adjusted) 100 years later!
I believe that is what the Schiller study also shows. If history can be used as an indicator, it would seem to be saying that housing prices in S.F. could stay the same in real value for decades.
Posted by: SchillerRocks! at September 28, 2006 4:24 PM
here is another link to the graph if the first one didn't work. (and yes I know its on a 'bubble' blog site but it is the only other link I could find)
Posted by: badlydrawnbear at September 28, 2006 4:29 PM
I’d like to clarify that my original question stems from an interesting point I heard on an NPR panel of which I only caught a portion. The speaker was stating that if you look at the last 20 years - average the appreciation (both the gains and the losses) and adjust it for inflation - the gains overall in the Bay Area are actually quite a bit more modest than most reports show. The fact that it’s mostly a useless number is not lost on me, but I would have liked to add it as another subtle shade in the overall picture I’m painting of the local economy because I hadn’t heard it before and it was interesting to me. I couldn’t remember what the actual number was, or who the speaker who said it was. I am assuming that no one else knows that number either, and by the sounds of it doesn’t think it matters anyway. My curiosity will just have to go unsatisfied.
JS, don’t you think historic real estate cycles are useless as well since population and the overall landscape of the globe is changing so dramatically, especially in the last 30-40 years? Loved the Graph Badly Drawn Bear, not sure what I take from it until I look at it more, but at least it has inflation adjusted pricing!
Posted by: I have a question at September 28, 2006 4:53 PM
I would have been interested as well to see what the local inflation adjusted gains were as well. Pity you cant remember who the speaker was.
Here's a chart I came across published by the FDIC. It shows historically boom & bust periods of real estate but only since the late '70's and only in select cities across the US
The full article and the FDIC's conclusions can be found here keeping in mind it's 3 years old and still doesn't help answer the bubble or not question, but it does show busts dont necessarily follow booms.
Posted by: anon at September 28, 2006 5:27 PM
Another post addressed the question of getting data back to 20 years. To calculate the last three years, which that aforementioned report does not cover, you need the multiple listing. County data tells you the sales price and date but to know if the market is rising or falling, you need to check the current sales, current listings and the expired listings. If you analyze enough of this data, you can calculate the appreciation rate. Keep in mind that appreciation not proportional across neighborhoods and it may move in fits and starts. I put a article on how to appraise property on my blog. If you are interested, click on my ame
Posted by: Paul James the Appraiser at September 14, 2009 11:59 PM