30+ Years of Housing Market Cycles in the San Francisco Bay Area

Note: Most of these charts generally apply to higher-priced Bay Area housing markets, such as those found in much of San Francisco, Marin, Central Contra Costa (Lamorinda & Diablo Valley) and San Mateo Counties. (Different market price segments had bubbles, crashes – or adjustments – and recoveries of differing magnitudes in the last cycle, which is addressed at the end of this report.)

Regardless of how recent cycles have played out, it is vital to understand how extremely difficult it is to predict, with any accuracy, when different parts of a cycle will begin or end. Boom times can go on much longer than expected, or get second winds; recessions or crashes can appear with startling suddenness. It should also be noted that all of the major down cycles in the Bay Area in recent decades have been tied to national or international economic factors, i.e. our cycles aren’t simply local events, separated from the rest of the state or country. However, it is true that local factors sometime exacerbate a downturn (the earthquake of 1989; our greater exposure to the dotcom bubble), or supercharge a recovery (the Bay Area high-tech boom of recent years).

Market Cycles: Simplified Overviews

Up, Down, Flat, Up, Down, Flat…(Repeat)

The chart below graphs ups and downs by percentage changes in home prices at each turning point.

Smoothing out the bumps delivers the simplified overview above for the past 30 years. Whatever the phase of the cycle, up or down, while it is going on people think it will last forever. Going up, it’s never going to stop! And then every time the market goes down, the consensus becomes that real estate will not recover for decades (or even “in our lifetimes”). But the economy mends, the population grows, people start families, inflation builds up over the years, and repressed demand of those who want to own their own homes builds up. In the early eighties, mid-nineties and in 2012, after about 4 years of a recessionary housing market, this repressed demand jumped back in (or “explodes” might be a good description) and prices started to rise again. (The dotcom bubble adjustment caused no lasting recession in home values.)

It’s not unusual for a big surge in values to occur in the first couple of years after a recovery begins, often quickly exceeding the previous peak value.

All bubbles are ultimately based on irrational exuberance and/or criminal behavior, whether exemplified by junk bonds, Savings & Loan frauds, dotcom stock hysteria, “Dow 30,000” insanity, “the end of the business cycle” nonsense, gorging on unsustainable debt, runaway greed (without any corresponding desire to produce anything of value), predatory lending, or dishonest financial engineering.

However, it should be noted that the most recent subprime-financing/ loan-fraud bubble was truly abnormal in its scale, and its crash was much greater than other “market adjustments” going back many decades. The pre-2008 bubble was fueled by tens of millions of buyers purchasing (or owners refinancing) homes with loans that they clearly couldn’t afford right from the get go: Liar loans, deceptive teaser rates, promises of non-stop appreciation, and the abysmal decline in underwriting standards. (Since lenders were simply selling the loans, they didn’t care about qualifying buyers anymore.) Sometimes there was no actual investment in the properties being bought, i.e. no down payment, 100%+ loans. Many lenders and mortgage brokers clearly engaged in criminally predatory behavior, convincing people to overload themselves with unsustainable, impossible-to-pay-back levels of debt. (Sadly, something similar has been going on in recent years with college loans.)

The market adjustments of the early 1990’s and that subsequent to the dotcom bubble saw declines in home values in the range of 10% to 11%, which is of a completely different scale from the recent 2008 – 2011 crash and decline, when values plunged by up to 60% around the state and country, depending on area and price segment. The previous crash of a similar magnitude was during the Great Depression of the 1930’s.

This is important context when contemplating the next adjustment: It doesn’t have to be a devastating crash. It can be more like some air being let out of an over-pressurized tire instead of a blowout on the highway at high speed. As of early-2017, it appears that the SF economy and housing market have cooled to some degree after 4 years of feverish appreciation. But the change varies by segment: The affordable house segment remains quite hot; more expensive house prices have generally plateaued; the condo market has cooled much more and seen price declines, and the luxury condo market has cooled the most. The condo market has been affected by the surge of new-construction condos hitting the market recently. We will have to wait and see the scale and speed of any further adjustment, but so far, we don’t see local or macro-economic conditions for a 2008-like crash. Adjustment, yes; devastating crash, no.

This Recovery vs. Previous Recoveries

The light blue columns in the above chart graph the home-value appreciation that occurred in the first three years of each recovery – our latest rebound has been somewhat quicker than other recoveries, probably due to 1) the depth of the previous market decline, and 2) the huge, high-tech employment, population and wealth boom that has played out in San Francisco and nearby counties. The gray columns chart the appreciation of past recoveries from the beginning to peak value for each cycle (except for the latest cycle, for which the peak has not yet been defined), and the red bars delineate the percentage declines from those peaks, pursuant to the market adjustments that occurred. As always, note that market appreciation and depreciation rates can vary widely by county, community and neighborhood.

Over the past 30+ years, the period between a recovery beginning and a bubble popping (or a lesser adjustment occurring) has run 5 to 7 years. We are currently about 5 years into the current recovery, which started in early 2012 (in San Francisco; later in outlying Bay Area counties). Periods of market recession/doldrums following the popping of a bubble have typically lasted about 3-4 years. (The 2001 dotcom bubble/ 9-11 crisis drop being the exception.) Generally speaking, within about 2-3 years of a new recovery commencing, previous peak values (i.e. those at the height of the previous bubble) are re-attained — among other reasons, there is the recapture of inflation during the doldrums years. In this current recovery, those homes hit hardest by the subprime loan crisis — typically housing at the lowest end of the price scale in the less affluent neighborhoods, which experienced by far the biggest bubble and biggest crash — are appreciating quickly now, but taking longer to re-attain peak values. However, communities with higher priced homes — such as in San Francisco, Marin, San Mateo and Central Contra Costa Counties (Diablo Valley & Lamorinda) — have surged well past their previous peaks.

This does not mean that these recently recurring time periods necessarily reflect some natural law in housing market cycles, or that they can be relied upon to predict the future. Real estate markets can be affected by a bewildering number of local, national and international economic, political and even natural-event factors that are exceedingly difficult or even impossible to predict with any accuracy.

In the 2 charts below tracking the S&P Case-Shiller Home Price Index for the 5-County San Francisco Metro Area, the data points refer to home values as a percentage of those in January 2000. January 2000 equals 100 on the trend line: 66 means prices were 66% of those in January 2000; 175 signifies prices 75% higher.

1983 through 1995 

(After Recession) Boom, Decline, Doldrums

In the above chart, the country is just coming out of the late seventies, early eighties recession featuring terrible inflation, stagnant economy (“stagflation”) and incredibly high interest rates (hitting 18%). As the economy recovered, the housing market started to appreciate and this surge in values began to accelerate deeper into the decade. Over 6 years, the market appreciated about 100%. Finally, the late eighties “Greed is good!” version of irrational exuberance — junk bonds, stock market swindles, the Savings & Loan implosion, as well as the late 1989 earthquake here in the Bay Area — ended the party.

Recession arrived, home prices sank about 11%, sales activity plunged and the market stayed basically flat for 4 to 5 years. Still, even after the decline, home values were 70% higher than when the boom began in 1984.

1996 to Present 

(After Recession) Boom, Bubble, Crash, Doldrums, Recovery

This next cycle looks similar but elongated. In 1996, after years of recession, the market suddenly took off and continued to accelerate til 2001. The dotcom bubble pop and September 2001 attacks created a market hiccup (a short-term 10% decline, but only for high-price tier houses, and for condos), but then the subprime and refinance insanity, degraded loan underwriting standards, mortgage securitization, and claims that real estate values never decline, super-charged a housing bubble. Overall, from 1996 to 2006/2008, the market went through an astounding period of appreciation. (Different areas hit peak values at times from 2006 to early 2008.) The air started to go out of some markets in 2006-2007, and in September 2008 came the financial markets crash.

Across the country, home values typically fell 20% to 60%, peak to bottom, depending on the area and how badly it was affected by foreclosures — most of San Francisco, with relatively few foreclosures, got off comparatively lightly with declines in the 15% to 25% range. The least affluent areas got hammered hardest by distressed sales and price declines; the most affluent were usually least affected. Then the market stayed flat for about 4 years, albeit with a few short-term fluctuations. Tied to a rapidly recovering economy, supply and demand dynamics began to significantly change in San Francisco in mid-2011, leading to the market recovery of 2012.

The Recovery since 2012 (Case-Shiller) 

This chart above looks specifically at home price appreciation since 2012 when the current market recovery began. Generally speaking, the spring selling seasons have seen the most dramatic surges in appreciation. It’s not unusual for appreciation to slow or flatten in the second half of the year. This chart below illustrates the connection between seasonality and appreciation over the past 4 years. The market in San Francisco was definitely cooler in Spring 2016 than in the previous 4 spring selling seasons, and we are now waiting to see how the Spring 2017 market develops.

Short-Term Trends by Price Segment/Property Type 

In late 2015 and 2016, the greatest pressure of buyer demand started moving to more affordable home segments, as seen in this following chart. The highest price tier has generally plateaued; condo prices appear to be declining with the surge of new-construction condo projects hitting the market; and the lowest priced tier continues to appreciate as buyers search for affordable housing options. But remember that short-term trends sometimes fluctuate without great meaningfulness.

The Panorama: From the late 1980’s to Present

S&P Case-Shiller Index, 5-County SF Metro Area

In the chart below showing percentage year-over-year changes, each January percentage change mostly reflects the market in the previous year, i.e. the January 2002 percentage decline reflects the change in 2001 after the dotcom bubble popped.

Comparing San Francisco vs. United States

Home Price Appreciation Trends since 1987 

Really quite similar except for the 1989 earthquake, the dotcom phenomenon, and the recent Bay Area high-tech boom. Of course, the huge difference is in the median house sales prices: The city’s is now over 5 times higher than the national median price.

FHFA Home Price Index

San Francisco & San Mateo Counties 

The Federal Housing Finance Agency also has its own home price index using repeat sales information on houses whose loans were purchased or securitized by Fannie Mae or Freddie Mac. Since the allowable loan thresholds are relatively low compared to SF house prices, it is not tracking the entire market, but it provides another angle on appreciation going back to 1975 (further than other data sources we have) – and its analysis generally parallels Case-Shiller and median sales price trends. The FHFA uses a metro area comprised of SF and San Mateo Counties.

San Francisco Median Sales Price Appreciation 

The charts below look at median sales price movements in San Francisco County itself over the shorter and longer terms. These do not correlate exactly with Case-Shiller – firstly because C-S tracks a “metro area” of 5 Bay Area counties, and secondly, because C-S uses its own proprietary algorithm and not median sales prices. Median sales prices are often affected by other factors besides changes in fair market value (such as significant changes in the distressed, luxury and new-construction market segments; seasonality; buyer profile; and so on).

The Current Recovery: 2012 – Present 

In 2011, San Francisco began to show signs of perking up. An improving economy, soaring rents, low interest rates and growing buyer demand coupled with a low inventory of listings began to put upward pressure on prices. In 2012, as in 1996, the market abruptly grew frenzied with competitive bidding. The city’s affluent neighborhoods led the recovery, and those considered particularly desirable by newly wealthy, high-tech workers showed the largest gains. However, virtually the entire city soon followed to experience similar rapid price appreciation.

San Francisco median home sales prices increased dramatically in 2012, 2013, 2014, and then again in the first half of 2015. In 2016, the SF market clearly cooled compared to the competitive frenzies of previous spring selling seasons, and home prices appear to be plateauing in year-over-year comparisons. But different markets within the city are experiencing different dynamics – the more affordable house segments, for example, are still extremely competitive. Again, you might want to review some of our most recent analyses by clicking on our Trends & Analysis link at the top of this webpage. As spring 2017 begins, initial indications point to a strong demand, low inventory dynamic – but it is too early to come to any definitive conclusions.

Median Sales Price Changes – Longer-Term: 1993 – Present 

In the chart above looking at overall median price changes for all SF house price segments, there wasn’t a decline after the dotcom bubble, but which does show up in the Case-Shiller charts for the high-price house segment and the condo segment. Overall median sales prices for condos, seen in the chart below, did fall in 2001. The condo segment (and apartment rents) seems more affected by changes in the high-tech economy than the overall house market.

Comparing San Francisco, California & National

Median Price Appreciation 

2012 through 2016, San Francisco has been out-performing the overall state and national markets.

San Francisco Rents

Besides, home prices, home rental rates are major indicators of what is occurring with housing costs and the local economy. If anything, rents have appreciated even more extremely than home prices in San Francisco (and other areas of the Bay Area) – and, of course, renters get no advantages from low interest rates, multiple tax deductions and advantages, or home-price appreciation over time. One classic indicator of an overpriced home market is when prices outpace rents. So far, this has not happened in San Francisco: Both types of housing costs have soared in recent years.

It’s interesting to note that SF rents actually dropped much further after the dotcom bubble burst than after the 2008 financial markets crash, though the latter was a much more destructive economic event. It suggests that local rents may be more affected by the simple ebb and flow of high-tech hiring and employment than by other macro-economic issues, such as stock market changes. If one loses one’s job and the likelihood of finding another in the area plunges, it may be an immediate imperative to move to a less expensive rental area (pressuring rents lower); if one’s net worth plunges with a stock market crash, one may no longer afford to buy a home (pressuring home prices lower). This is an oversimplification, but may still go some ways to explaining the different scale of reaction by purchase and rental markets to different macro-economic events.

As of mid-2016, the SF rental market has definitely cooled, with supply increasing significantly with new construction, demand softening, and rents beginning to decline, especially at the high end. According the the latest data, as of Q1 2017, SF average asking rents have dropped around 8 – 10% from their peaks in 2015.

Consumer Confidence

The monthly fluctuations in consumer confidence reported on in the media are relatively meaningless and without context, but longer-term movements are much more meaningful to overall economic trends. Psychology – confidence, optimism, fear, pessimism – often plays a huge role in financial and real estate markets. And events can sometimes turn consumer confidence one way or another very rapidly, whether such movements are rational or not.

Mortgage Interest Rates since 1981

It’s much harder to decipher any cycles in 30-year mortgage rates. Rates remain very low by any historical measure, but have risen since the 2016 election. Interest rates play a huge role in the ongoing cost of homeownership (affordability) and the real estate market. The substantial decline in interest rates since 2007 has in effect subsidized much of the price increases that have occurred since 2011.

Employment Trends 

Real estate market cycles have a symbiotic relationship to other economic cycles, such as illustrated in the employment charts above.

Housing Affordability by U.S. Metro Statistical Area per National Association of Realtors 

Housing Affordability Index (HAI) Cycles, 1991 – Present by Bay Area County, per CA Association of Realtors

Unsurprisingly, there is a reverse correlation between the trend lines for housing affordability rates and those of real estate price cycles (above). HAI rates jump higher in market recessions, peaking at the bottom of the market, and then decline as the market recovers, bottoming out when peak prices are hit. The lowest Bay Area housing affordability housing index rates (probably in history) were hit in 2007 right before the 2008 market crash. The Bay Area overall is still above those lows in its current recovery.

The 2008 San Francisco Bay Area real estate crash was not caused just by a local affordability crisis: It was triggered by macro-economic events in financial markets which affected real estate markets across the country. It is important to note that in the past (certainly going back at least 50 years), major corrections to Bay Area home prices did not occur in isolation, but parallel to national economic events (though the 1989 earthquake, which occurred just before the national recession began, certainly exacerbated the local downturn). Ongoing speculation on local bubbles (and predictions of awful upcoming local crashes) often neglect to remember this.

Still, dwindling affordability is certainly a symptom of overheating, of a market being pushed perhaps too high. Looking at the chart above, it is interesting to note that the markets of all Bay Area counties hit similar and historic lows at previous market peaks in 2006-2007, i.e. the pressure that began in the San Francisco market spread out to pressurize surrounding markets until all the areas bottomed out in affordability. This suggests that one factor or symptom of a correction, is not just a feverish San Francisco market, but that buyers cannot find affordable options anywhere in the area. We are certainly seeing that radiating pressure on home prices occurring now, starting in San Francisco and San Mateo (Silicon Valley) and surging out to all points of the compass.

San Francisco’s Housing Affordability Index (HAI) has been running about 3%-5% above its all-time historic low in Q3 2007, but affordability in most other Bay Area counties, while generally declining, still remain significantly above their previous lows. By this measure, the situation we saw in 2007-2008 has not yet been replicated.

Significant increases in mortgage interest rates would affect affordability quickly and dramatically, as interest rates along with, of course, housing prices and household incomes, play the dominant roles in this calculation.

Inflation & Interested Rate-Adjusted Housing Cost (since 1993) 

The Home Cost Trends chart below (a little out of date as of early 2017) reflects a very approximate calculation of monthly home payment costs (principal, interest, property tax and insurance) adjusted for inflation, i.e. in 1993 dollars, using annual median house sales prices, average annual 30-year interest rates, and assuming a 20% downpayment. The average annual compounding CPI inflation rate fluctuated, but averaged approximately 2.4% over the period, and average annual mortgage rates fluctuated from 8.4% to 3.7% (see mortgage interest rate charts earlier in this report), which, as mentioned before, had a huge impact on financing costs.

Adjusting for inflation and interest rate changes means that though the median sales price is now far above that of 2007, the monthly housing cost is still a little bit below then. This isn’t a perfect apples-to-apples comparison because it doesn’t take into account that the amount of the 20% downpayment increased significantly over the time period. Still, since ongoing cost is typically an important factor for homebuyers (at least those getting financing), this affords another angle on our market.

Different Bay Area Market Segments:

Different Bubbles, Crashes & Recoveries 

The comparison composite chart dramatically illustrates the radically different market movements of different Bay Area housing price segments since 2000. Farther below are updated individual price charts for each price segment.

Again, all numbers in the Case-Shiller chart relate to a January 2000 value of 100: A reading of 220 signifies a home value 120% above that of January 2000. The chart above illustrate how different market segments in the 5-county SF metro area had bubbles, crashes and now recoveries of enormously different magnitudes, mostly depending on the impact of subprime lending. The lower the price range, the bigger the bubble and crash. In the city itself, where many of our home sales would constitute an ultra-high price segment, if Case-Shiller broke it out, many of our neighborhoods have risen to new peak values. The lowest price segment, more pre v a lent in other counties, may not recover peak values for some time to come. Updated C-S charts for each price segment are below.

Since mid-2016, the low-price tier has begun taking the lead in home price appreciation (though, again, it remains below its previous peak value).

Updated Case-Shiller Price-Tier Charts

Low-Price Tier Homes: Under $620,000 as of 11/16 

Huge subprime bubble (170% appreciation, 2000 – 2006) & huge crash (60% decline, 2008 – 2011). Strong recovery but still below 2006-07 peak values.

Mid-Price Tier Homes: $620,00 to $985,000 as of 11/16

Smaller bubble (119% appreciation, 2000 – 2006) and crash (42% decline) than low-price tier. Strong recovery has put it over its 2006 peak.

It is interesting to note that the low and mid-price house tiers basically shrugged off the dotcom bubble popping in 2001, while the high-price house tier and condos (and apartment rents) saw significant declines. This is another example of how difficult it can be to make big, general pronouncements regarding the entire Bay Area market. center>

High-Price Tier Homes: Over $985,000 as of 11/16

84% appreciation, 2000 – 2007, and 25% decline, peak to bottom. Now well above previous 2007 peak values.

Bay Area Condo Values 

After a strong recovery, recently seeing a dip in median sales prices, estimated in San Francisco itself (as opposed to the 5-county metro area) to be in the 4% – 5% range over the past year.

20170900