patrick.net

The last time the economy was this bad (1982) we threw everything we had at it to get out:

Unemployment (blue), Fed rate (red), Debt to GDP (Green, left scale)

The story this chart tells is pretty easy to see. Volcker kills the economy by pushing fed rate to near 20% in 1981, debt to GDP declines in response, and big double-dip recession with unemployment moving from 7.5% to 10%. Thanks, Paul.

But starting in 1981, debt to GDP was allowed to rise from ~1.0 to 1.5. Starting in 1982 interest rates fell from 15% to under 7.5%, and unemployment fell back to ~6%.

Today, however, we find ourselves in a very different situation.

Same graph, 1998-2010

The blue unemployment line is pegged at ~10%. It'd be higher but "participation" is falling and we've had 99 weeks of unemployment to fake-employ people, good for another 4% better number.

Fiscal policy (the red line) is clearly at the zero-bound, so the Fed is reduced to printing money, which will probably inflate everything EXCEPT people's salaries.

So we're left with debt as our savior this decade. It's already well above 2.0 x GDP.

Treasury is doing its part:

http://research.stlouisfed.org/fred2/series/FYGFDPUN

But everyone else seems to be tapped out:

http://research.stlouisfed.org/fred2/series/TCMDO

I'm not going to downplay this rise in foreclosures but I will say it's only part of the story.

The other part being that the bay area is diverse. In my local market that I watch very closely Case Shiller index will
be going up this spring. No question about it.
Rents also going up too.

@Troy

Thanks for the graphs... Now to look at them again...

If this downturn stretches up to 2025, would it be considered bad then Great Depression?

The misconception that low interest rates (i.e. 4%) is in any way part a bottom/buying opportunity is the kind of true idiot thinking that got everyone into this mess. Nothing would kill real estate prices quicker than a rate increase to 6% (a number we were told was a "historic low" only a few years ago). As Patrick has pointed out in other postings, for those who have actual savings toward a home purchase it would be far better to buy when rates were high and prices low. Right now we still have the opposite. Come on Chinese! Bring on the 7-8% rates and let's get on with it.

Very nice chart YesYNot. Could you plot the curves for the fed funds and the rate of change for the median home price as a moving average to smooth out the noise a bit?

Sorry, I should have stated this earlier. In the first plots, the FFR data were monthly, and smooth enough, so no averaging was done. The median price data had a lot of monthly fluctuation, so the derivative was meaningless. In the first plot, I did a 13 point moving average of the monthly data, essentially averaging over the year. It looked smooth enough, so I left it alone.

In the next plot (this post), I did a 11 point moving average on the fed funds rate & the already averaged median price data. This one is a lot smoother, and the lines have not otherwise changed much. As expected, the averaging made the peaks a little less extreme. For instance, the peak interest rate went from 19% to 17% or so in 1981. Applying the averaging twice like this allows data 12 places away 1 yr in either direction to influence the value at any one point.

The derivative is center weighted: (price_n+1 - price_n-1)/(2/12) := $ / year. This was normalized to percent change by dividing multiplying by 1/(price_n)*100.

Thanks, the trend looks a lot cleaner, particularly for the fed fund peaks around 1970, 1980, 1988, and 2006. The only strong exception seems to be ~1972-1976 where the fed funds and the price rate-of-change look to correspond. This also corresponds with the beginning of your trend line for Freddie Mac 30-yr in your earlier chart. Any relationship between the two?

Also, the rate-of-change response to the fed fund rate in the 1990s seems a bit weak relative to other peaks...

SF ace says

An accompany rise in the fed fund rates after the crash (and ultimetly rise in interest rates) lead to apppreciation subsequently.

Another interesting note about the graphs is that a double dip is extremely unlikely. There was negative appreciation 3 times - around 1970, 1990 and and late 2000's - dips ~every 20 years. Maybe this time will be different though.

YesYNot says

This seems to me to be great evidence that increased borrowing costs slow down demand.

It's utterly bizarre why this is even being debated.

The core miscommunication here is an expectation of late 1970s, late 80s, or late 90s style job market heat-up.

While there is an obvious pattern there, one needs to understand WHY these booms came as they did.

I think the 1970s was a bona-fide productivity boom, driven by the baby boom, women, and available opportunities.

The 1980s continued this but started the divergence of total debt vs. GDP.

Total Credit Money Owed (blue) vs nominal GDP (red)

Plus lower nominal oil prices 1985-1995 didn't hurt.

The 1990s featured the front-side benefits of our burgeoning trade imbalance, largely with China:

http://research.stlouisfed.org/fred2/series/BOPGTB

So yeah, if we get a job boom like previous decades:

Job growth in 5 year periods:
1970-1975: +8.6%
1975-1980: +17.5%
1980-1985: +6.1%
1985-1990: +13.3%
1990-1995: +6.6%
1995-2000: +12.4%
2000-2005: +1.3%
2005-2010: -2.4%

we'll see home inflation like those hot periods (1975, 1985, 1995).

Right now we're at the same number of jobs we had back in 1999, so any growth now is just bailing out the boat.

However, debt -- per the graph above -- has shot off the chart.

And now we've got 80 million baby boomers ready to enjoy their trillions of government cheese they've been promised.

This is not going to end well.

Here's a chart I like to make about population pyramids . . . you can easily compare certain years since the additive blend of their year cohorts turns gray thanks to the bar being complementary colors.

With this chart solid color is new additions to each 5 year group, 2020 vs 2003.

Eg. in 2020 we will have about 3 million more people age 80+ in 2020 compared to 2003.

You can really see the baby boom hit the 55-75 bins, too. Roughly 18M more people up there by 2020.

This is the demographic trend we're looking at. Inflation is not going to solve it.

Curiously, people in their 40s will fall by 4M people in 2020. I don't know what this means.

fwiw, here's the Japanese comparison, 2003 vs 2020:

It's kinda unclear but other than the seniors, only the 40-49 cohort is larger in 2020 than 2003.

The age 25-35 cohort is going to be missing ~8 million people!

I don't know if the Japanese pyramid is a lot worse or a lot better than ours.

The number of old people 80+ doubling from 5.5M to 11M is something, but I don't see why their unemployment problem won't go away later this decade.

Except for people in industries that cater to young adults I guess. That cohort is just disappearing.

Troy,

The charts are interesting.

When I look at it [and make the assumption that boomers will be retiring at "normal" ages], it appears that we will be removing far more people from the labor force than we are adding. (The yellow part of the 65-69 bar is much larger than the blue part of the 20-24 bar)....so, with time, our unemployment situation will get some help, I imagine. I think Japan is going to be hurting pretty bad over the next 3 decades or so -- especially if people just keep getting older instead of dying. :)