Sunday, April 24, 2016

Another Global Housing Bubble?

In 2007, excessive debt led to a housing boom and bust of extraordinary proportions. Household debt in a handful of countries peaked in 2007, at the height of the boom. The countries (ex. US, UK, Spain) that went through the crisis have somewhat de-levered their household balance sheets. Household in many other countries, unscathed by the crisis, have continued their relentless borrowing.
For example, Australia, Canada, Norway and Netherlands have household debts far higher than the U.S. at its peak. There are likely structural differences for the different levels of debt such as regulations, lending practices, interest rates, etc. But how much is too much? At some point too much debt will increase the risk to the entire financial system. One argument for higher debt is that interest rates remain low. The servicing costs of debt is relatively stable despite much higher borrowings. On the other hand, household debt is typically used for consumption or to buy a house, not the most productive use of capital given the low returns on rent. As more and more money gets tied into illiquid, unproductive assets, who ends up paying off the debt? Will the principal ever be repaid?

Even as the world recovery remains tepid, the housing markets in many of these countries caught on fire, fueled by debt-driven demand. Take Canada, a country whose economy is closely associated with the U.S.. With commodity prices facing severe overcapacity and weak export demand, the country's GDP growth has been weak. Yet, housing prices in the country's two main cities, Toronto and Vancouver, have been extremely strong with prices growing by double-digits year-over-year.




Foreign demand has certainly played a role in driving up prices. Yet, it's probably disingenuous to attribute all of the price action to foreign capital. Just looking at the base rate, with millions of households in these two cities, the prime buyers of Vancouver and Toronto real estate remain the residents of these cities. But how can locals afford ever increasing prices with relatively stagnant wages? Hint: credit growth. Consumer in Canada has continued to grow at around 6% annually, or double the rate of GDP growth mainly due to lower interest rates. As such, the carrying costs of owning a house in these two regions have been relatively flat despite the higher debt levels. The question remains, unless wages pick up significantly, how will people afford to repay the debt?


I suspect that, most probably won't. The repayment of debt will likely be shared by society. At some point, too much debt changes from a household problem to a systemic problem. (They will in the traditional sense, but because their wages got a boost from higher inflation.)

In Canada, around 50% of mortgages (mostly high LTV) are insured by the CMHC, a government entity. Mechanism like these help de-risk the system in the event of a financial crisis. However, it means that the costs of a potential debt crisis will be borne by all members of society.

The fall-out will likely be even lower rates and a depreciation of the currency, which decreases real purchasing power so that nominal wages can stay flat or rise moderately. Or higher inflation to boost nominal wages, even as real purchasing power stays flat, which will erode the value of liabilities over time.

This leads to a dilemma for the investor. Do you remain financially prudent even though you will borne the costs of society's excessive leverage, or do you become financially imprudent to try and benefit from the eventual fallout?

4 comments:

  1. Do you think CMHC will nationalize the debt? If not, I'd rather be financially prudent.

    ReplyDelete
    Replies
    1. In a sense, 50% of the debt in Canada is "nationalized". (Not in the sense that they own all the debt, but that they're on the hook for defaults.) The government backstopping of debt creates a mortgage market for higher risk borrowers that would not have existed. It's much easier for lenders when they do not have credit risk.

      This is not a problem when debt levels are low. Premiums can absorb losses and refinancings at lower-interest rates will help most borrowers. However, as debt levels rise, there's a level when it moves beyond an individual's problem to an economic crisis. In such an instance, any losses in excess of CMHC's capital will have to be covered by the government.

      I think the bigger problem is what happens when credit slows. When there's too much debt in the economy, the effectiveness of continued borrowing decreases. (ex. The "velocity" of money decreases.) To create growth, you need to either reduce debt so that people can borrow again or create inflation. Governments tend to choose the latter path. The easiest way to reduce debt to GDP is for the government to issue a cheque to every person.

      Essentially, borrowers get bailed out and savers gets to pay. Greece and Puerto Rico may provide interesting examples of how things play out.

      Delete
    2. That is possible and would not shock me if that happened. Do you think Canada will have a housing crisis? Price increases are clearly detached from reality.

      Delete
    3. I don't know. It's hard to say for sure. As prices go up, the possibility of a crisis increases, but it's by no means guaranteed.

      I think most housing bears miss a scenario where interest go even lower. Most bears seem to believe, with certainty, rates will definitely go up, which will cause a crash.

      However, based on what's happened to interest rates over the past 30 years, it's more likely that rates will drop, potentially into the negatives, than go up. So there is a distinct possibility that Canada can avoid a housing crisis by lowering interest rates even further. Imagine refinancing at 0% mortgage rates and how that will help sustain the market.

      I mean, between the government's two choices of 1) higher rates, lower housing prices, higher unemployment, higher C$ or 2) lower rates, higher housing prices, lower unemployment, lower C$, most would probably skew toward 2).

      Delete