Disaster myopia: floods and financial crises

Apart from the tragic loss of life, one of the saddest outcomes of the Queensland floods is the financial ruin that hundreds if not thousands of families are now facing having had their properties destroyed. Many do not have flood insurance. This recent article in The Age tells one typical story.

Landlords across Queensland are likely to go bankrupt in the next few months, an industry expert has warned...

Paul and Sarah Smith had been renting Sarah’s mother’s rental property at Goodna when the floods hit. Their home was inundated to the second story and the damage is enormous.

Ms Smith said her mother owned five other properties in Goodna which she rented out – all of them had gone under in the flood last week.

“My mum has six rental houses and they’re all ruined,” Ms Smith said.

“You might say ‘oh she’s got six rental properties, she must be rich’, but she’s not – she has mortgages on all of them and relies on the rent to help pay.

“How will she pay for them now? She’s devastated. She doesn’t have insurance for this...”

Now, I have nothing but sympathy for the people involved. But you do have to ask the question. How sensible is it to have your entire net wealth tied up in six rental properties that are all geographically concentrated in a single flood prone area? And on top of that, to not bother getting flood insurance?

Disaster Myopia 
This is a classic example of what economists and psychologists call "disaster myopia" -- the tendency of people to greatly underestimate the chance of catastrophic events. Here's how it works. You get a big disaster, say the 1974 floods in Queensland. This serves as a wake up call for people that bad things can and do happen. For a while, they start to behave more prudently: more people take out flood insurance, builders become wary of building in vulnerable areas, and banks start to take into account the flood vulnerability of houses before they issue mortgages. But sooner or later, memories of the disaster start to fade. As the years pass, fewer and fewer people bother getting flood insurance. Developments start cropping up closer and closer to river banks. Banks stop worrying about flood risk and issue loans to whoever is willing. We all get a bit complacent. And then another disaster strikes...

On a broader scale, this phenomenon is exactly what has got much of the global economy in such a big mess since 2008. In fact, the risky behaviour by banks that was at the heart of the recent global financial crisis can basically be seen as the result of disaster myopia, Andrew Haldane of the Bank of England has argued. After a "golden decade" of high economic growth, rising house prices and low inflation around most of the world in the 2000s, banks started lowering their lending standards, based on the idea that the world had entered a new phase of economic stability. After all, there hadn't been a major global financial shock for about a decade, the period of time that seems to mark the limits of the attention span of risk managers in banks, Haldane notes:

As time passes, convincing the crowds that you are not naked becomes progressively easier. It is perhaps no coincidence that the last three truly systemic crises – October 1987, August 1998, and the credit crunch which commenced in 2007 – were roughly separated by a decade.

The problem is, the "golden decade" that banks were basing their decision making on, was a highly unusual one compared to history, as the two displays of UK unemployment and corporate earnings growth show below.

Source: Bank of England

Apologies if this is bringing back nightmares of Statistics 101, but you can see here that the probability distributions for both of these indicators looked very different in the "golden decade" to the longer-term history: in statistical terms, there was not a lot of variance around the mean. What did this lead to in practice?

If you were a banker with a short memory and only looked at the yellow curve, you might have concluded that property lending was very safe, because unemployment was "permanently" low and therefore the risk of borrowers defaulting on their loans was negligible. And bankers have an incentive to have short memories. After all, if everything blows up in 10 years because of all the stupid loans you made, who cares, because you still get to keep all the big bonuses you've made in the meantime.

And if you were an investor looking at the yellow curves above, you might have concluded that stocks were a very safe investment, because corporate earnings were growing steadily with very little volatility. This illusion of stability led to an underpricing of risk: people took on a lot of debt and pushed asset prices higher and higher. And then the US subprime crisis hit and markets all over the world came crashing down.

ANZ's Gold Medal
It's time to return this discussion to Australia, where disaster myopia is almost an art form in some circles. A wonderful example of this came from a report by ANZ last week, which was their latest attempt to justify the sky high level of Australian property prices. I'm not going to examine it in detail, because other bloggers have already done an excellent job at ripping their arguments to shreds (see here for example). However, I do want to pick up on the paragraph below from this summary of ANZ's view.

A new report from ANZ claims residential properties aren't overvalued and has taken aim at traditional methods of evaluating affordability, including price-to-income ratios, saying they don't take into account more complicated and less-quantifiable factors including historic declines in interest rates... More importantly, (ANZ economist) Montalti says, is how the market is performing now and where those prices are actually moving. He points out the country has quite a low delinquency rate, which is evidence of a largely affordable market that is valued correctly.

In this single paragraph, ANZ gives us two classic examples of disaster myopia.

Let's deal with ANZ's second point first. Montalti points to the low level of delinquencies in Australia and says that this is evidence that the market is valued correctly. With all due respect to Mr Montalti, this is one of the dumbest arguments I have heard in a long time. It's a bit like if a meteorologist had told you at the end of 2010 that there hasn't been a major flood in Queensland for 30 years, so there was "evidence" to prove that nobody ever needs to worry about the risk of floods anymore.  

In any case, back to ANZ's first point, which is almost as silly. What they're telling us here is that the very high level of house prices in Australia is justified because the historical decline in interest rates (in one previous report they described this decline as "permanent") allows people to service a much higher amount of debt than in the past. And they're not the only ones making this argument. The IMF recently cited "permanently lower nominal interest rates that have occurred since 2000" in its analysis of Australian house prices.

But what if the low level of interest rates we've seen in Australia over the past decade isn't permanent? In fact, it would be very surprising if it was anything more than temporary. Below, you can see that I've charted the probability distribution of Australian interest rates along the lines of the Bank of England charts above. (I picked the 90-day rate because the RBA provides data for this one all the way back to 1969)

The result is quite interesting. If you only looked at the past decade, represented by the grey bars, you would see a nice bell-shaped looking distribution with interest rates generally clustered in the 5-7% range. But the red line, which shows the distribution of interest rates from 1969 to today, shows you that the past decade has been a bit of an anomaly in historical terms. The distribution is heavily skewed to the right.

Does this mean that interest rates are going to go back to 15% like we had in the 1970s? No it doesn't. But what it does suggest is that if you think the relatively stable level of interest rates we've had in the past decade is some kind of "new normal", you're probably in for a rude shock at some point in the future. And that's a bit of a problem, because Australian house prices today are so high that many borrowers are already forced to devote an uncomfortably high portion of their disposable incomes to mortgage payments even with rates at current levels.

How many people could deal with a rise in mortgage rates to 10%? Again, not to say that this is likely, simply that it wouldn't be all that out of the ordinary in historical terms.

A margin of safety
It might sound like the tone of this post is overly pessimistic. After all, if we became totally paranoid about all the possible disasters that could happen none of us would even leave the house every day. And we wouldn't take any kind of financial risks at all. But that's not really the point. The point is that when making financial decisions, we should allow for the fact that some of the assumptions we have about the world could quite possibly turn out to be wrong. For example, ideas like:

• Property prices in Australia double every seven years
• Unemployment under 5% is the norm for Australia
• The terms of trade boost from China will last for decades
• Interest rates will never go as high as 10% again

And so on and so on. Now, we can't possibly predict which of these risks we need to worry about (and we certainly can't rely on the likes of ANZ to tell us), but that doesn't have to matter so much if we incorporate a margin of safety into our decision making. Do you have enough of a buffer in savings so that you could keep paying your mortgage even if you were to lose your job and be unemployed for a few months? Could you keep paying your mortgage even if your mortgage rate went up to 9 or even 10%? If the answer to these questions is no, then perhaps you are living a little too close to the edge.

Oh, by the way, I didn't mention the risk that the earth might get hit by a massive asteroid in 2014, but I can assure you there's nothing to worry about there.

Until next time.