2017 economic headwinds: housing bubbles popping up, and popping, everywhere

Housing was the first of the 2017 headwinds set out on this site, because it was the one showing the most obvious signs of a top, in three countries at once. The material accumulated faster than it could be written, so the housing piece was broken out on its own. This page keeps the American case, the Canadian and Australian cases in brief, and the demographic argument that made the coming bust unlike the last one.

The United States: back to 2006 by the same road

The effort to return house prices to their 2006 peak had run for about seven years and by early 2017 had succeeded in most metropolitan areas. The peak had been reached the first time only through loose credit, since prices had gone beyond what incomes could support, and real wages in 2017 were no higher than they had been in the crash. So the second ascent required the same method. The Federal Reserve’s survey of senior loan officers showed that for eleven consecutive quarters more lenders had loosened mortgage standards than tightened them: lower minimum credit scores, lighter documentation for the self-employed, higher loan-to-value limits. The new administration’s economic adviser said on television that its executive orders were intended to relieve the scrutiny post-crisis regulation had put on banks, and the president had promised to do “a big number” on Dodd-Frank. The stated reason was that banks were not lending enough. Bank lending to business was at a record.

The signs of a top were the ones seen in 2006. Existing-home sales in January ran at a pace not seen since 2007. Houses stayed on the market fifty days nationally and under a month on the West Coast. Inventory was the lowest since 1999, and bidding wars over list price were back in Seattle and San Francisco. House flipping had returned to its previous-bubble level. And at the top of the market the fall had begun: Miami condo speculators who bought pre-construction were taking losses on completion, Manhattan luxury contracts were down a quarter and closings down 18.6 per cent year on year, with new-development inventory up 27 per cent. Barry Sternlicht of Starwood called the Manhattan luxury market a catastrophe that would get worse, and since that segment is bought with Wall Street bonuses, the question was what it said about Wall Street.

The demographic bubble

The argument that distinguished this piece from an ordinary bubble call was about who would buy. The baby boom, which had driven housing demand for forty years, was leaving the market. For the next two decades the fastest-growing age group would be people over seventy, whose numbers would roughly quadruple, and who buy about seven per cent of homes and cannot easily get a thirty-year mortgage. Every previous bust had been followed by a recovery driven by a larger cohort of buyers. This one would be followed by a smaller one. The observation about immigration followed from that: the second reason business wanted high immigration, after cheap labour, was that population growth is the only reliable way to grow a housing-based economy, and an administration restricting immigration was, whatever its intentions, removing the buyers at the bottom and middle of the market while the top was already sliding.

Canada and Australia

Canada’s bubble was more precarious than America’s and already falling in 2017: Toronto and Vancouver prices had risen far faster than incomes, a foreign-buyer tax in Vancouver had halted the market there, and Canadian household debt relative to income was the highest in the developed world. Australia was further along still. UBS had reported that a large share of mortgage applications contained misstatements, so-called liar loans, and a parliamentary inquiry had heard that white-collar misconduct at the banks was rife and unchecked. Property funds had frozen redemptions. The Sydney Morning Herald described a five-speed market in which the outer suburbs were already going backwards.

Where it stood

The claim was not that a crash was certain in 2017. It was that the conditions of 2006 had been recreated deliberately, in three countries, on the same credit, with a smaller generation of buyers behind them, and that the top end had already turned. A housing market at that point does not need a shock to fall. It needs only for the credit to stop loosening.

Later note. The credit did not stop loosening; it got cheaper. American prices rose another 50 per cent by 2022, as mortgage rates fell below 3 per cent in the pandemic, before the sharpest rate rise in forty years froze the market rather than crashing it. Canada’s and Australia’s prices fell in 2022 and 2023 and then resumed. The demographic argument is still in front of all three countries, one decade closer.