The black arrows in 1979 and 1989 indicate real estate bubbles. But they were just little bitty things compared to the whopping divergence between income and housing prices we have now.
So, onto the primary question, “what would be required to bring house prices and income gains back in line?”
The answer to that is either:1) An income gain of 51%
2) Or a decline in house prices of 34%
The chart is not particularly helpful for a number of reasons.
First, there is no market segmentation. It is unlikely that the prices are all across the board. If the aggregate figures are driven, for example, by the upper class members who have been benefiting from 7 years of Bushonomics, then the figures may not, in fact, be out of line. I don’t think this is true, but it’s impossible to tell from the chart.
Second, there are significant institutional changes: 40- and 50-year mortgages, no downs, etc., which might account for some of the discrepancies. Whether such financial instruments are wise over the long run is another question, but if people want to consume more now than in the future, so be it.
Third, there is the possibility of transfer payments, ie, draining savings to buy a house, or upgrading from one house to another and using the accumulated equity to buy in. The relative liquidity until recently of the mortgage markets has not led to the problems that existed in the late 1970’s when seller’s routinely carried back paper or engaged in wraparound loans that covered a huge portion of the sales price and endangered equity. Inheritance and savings (good investments from before the tech meltdown?) might also account for influxes. I don’t think they do, but I can’t disprove it from this simplistic graph.
All this being said, it’s clear to me, as it has been for the past couple of years, that the market will fall in due course because there is not a broad-enough base to sustain the runup in prices. THere’s clearly been a lot of panic of late in the markets leading to a final runup in prices before the collapse. But this graph doesn’t have nearly enough information to prove it.