by Tyler Cowen
on August 31, 2012 at 6:47 am
There is more here, from Sober Look.
What does the “US Calculation” rider alert to?
Wherever monetary policy, as conventionally understood, succeeds, it sows the seeds of its own failure.
Inflation targeting (and globalization) keeps wages and incomes low. Stimulative asset purchases keeps consumption and employment high. Debt rises, before the fall.
When the commodity ‘supercycle’ inevitably ends, Canada and Australia will go down too. Perhaps Norway as well, whee household debt to income and house prices to income is even worse.
“globalization) keeps wages and incomes low”
Just as long as you make sure never to count those pesky individuals that globalization gets a shiny new job.
So on this chart, a higher number means more debt? This chart is showing us that Canada has a higher debt to income ration than the US and has since about 2008? Is that right?
That’s right. We Canadians tend to be kind of smug about avoiding the banking excesses that started the crash in the U.S., but at least the U.S. has had four years of deleveraging since then… the housing bubble is still inflating in Canada. I have friends trying to buy houses and it’s unbelievably expensive in Toronto and the inner suburbs. At least student debt isn’t so rampant.
Wow – the debt to personal disposable income ratio is greater than 100:1?
Oh, wait, it’s just another useless graph with unlabeled axes. Let’s all speculate what they were trying to say and then draw conclusions from it.
Ratio versus Percentages.
In the article, the lead-in to the graph describes it as the “percentage” of debt compared to disposable income.
LOL The title of the graph tells you what the y-axis represents. Debt-to-income is a very common ratio in economics. Like debt-to-gdp when talking about sovereign debt. Writing “ratio” would be redundant.
Which is what makes the numbers in the y-axis hard to believe. I just ran the numbers for myself, using 8 months of income and counting my 401(k) contributions as tax (i.e., not included in disposable income). I have a mortgage, car loan and student loans. I calculated my ratio of debt: disposable income as just under 5 (for 12 months income, it would be around 3.5). Even throwing in my per capita share of the federal debt only gets me above 5.2 based on 8 months income.
Do I understand correctly that you would be way off the top of the chart? For example, you have $500,000 in debt ($400,000 mortgage, $50,000 car loan, $50,000 student loan) and $100,000 in annual disposable income? Am I missing something?
Tee-Hee!! Norman forgot to consider his decimal places!! You are correct Finch.
I wasn’t trying for a zinger; I might have missed something in Norman’s comment, or in the chart. I apologize if it sounded like I was.
My understanding is that people often start adulthood with high debt-to-income ratios and that they generally fall with time. It’s hard to buy a house in many places without a mortgage that exceeds twice your annual income. 5x, if I interpreted you correctly, does sound quite high.
Wouldn’t you subtract the value of the house from the mortgage, and the value of the car from the car loan? Seems we would be interested in net debt, not gross debt.
No, that was the essence of what I was saying. There are no decimal points in the y-axis to the graph, unless they are too small for me to see. As I noted below, in the article, the lead-in refers to debt to income as being a percentage, but the graph itself refers to ratio. Either one strikes me as being off (although as a ratio, it’s way off), since I am probably not all that unusual among the class of people carrying a mortgage.
The usual mortgage ratio is 2-3X gross income, not disposable income.
Yeah, then I _think_ the way to interpret this chart is that most people have a lot less debt than you relative to their income. I don’t think the usual mortgage is that high. It might be in Boston, say, but it’s not in Cleveland. And it might be for 28 year-olds, but it’s not for 50 year-olds.
FWIW, I’m at about 2x, or 200 on that graph, almost entirely because of my mortgage.
I agree with the many other commenters pointing out that the darn thing could be better labeled. Maybe there’s more explanation in the article I conspicuously didn’t read.
Yes, the graph should not say ratio, since it really seems to be the percentage (I found another source saying the average US debt is 120% of disposable income, which more or less matches the graph). The article seemed to think that Canada is facing a credit bubble that could burst like in the US. I took some comfort if the average person is less indebted than I am. The problem with the statistic, as many other commentators are pointing out, is that not all debt is the same. I am above the debt average solely because of my mortgage, but because that debt is secured by the house (and not my income in any meaningful sense because I have a lot of equity in the house), measuring that debt against my income is misleading. The debt to income ratio should be meauring debt that is “secured” by income, and even that is somewhat misleading: I could pay off my car loan and student loans tomorrow if I wanted from other liquid assets. Thus, in a real economic sense, my debt to income ratio is 0 since I am not dependent on current disposable income to pay my debts, but none of that is captured by the statistic, which indicates that my ratio is 3.5.
I don’t know. I think the reason higher percentages are sometimes appropriate is that a skilled young person might have a reasonable expectation that they will be employed for the next 30 years, while an unskilled 60 year old might not. So the skilled young person can reasonably expect to pay a higher debt service.
But it’s not the case that just being able to net out your assets and be left with something makes it all okay. If the 60 year old lost their job, being able to sell the house and car and not strictly speaking needing to declare bankruptcy might be cold comfort. It’s being young and high earning that makes it okay, or less bad.
I don’t think having a positive net worth means that debt is meaningless. That should be one of the lessons of the recent financial crisis. There’s value in flexibility. So I don’t think the graph is that misleading, labeling complaints aside.
If you click through to the source article you’ll see the metric excludes mortgages from individual debt.
> If you click through to the source article you’ll see the metric excludes mortgages from individual debt.
I don’t think so. I think the number quoted as excluding mortgages is in another article and another source, and is not referring to the graph. If the graph does exclude mortgages, it represents debt levels much higher than I think are typical.
But again, it’s not well labelled. Advice to graph makers: Label the graph. Include whatever notes are necessary to replicate the graph directly under the graph or in the title.
Fair point, the chart seems like it was created with Reuters EcoWin Pro which looks to be some type of charting tool. Unfortunately the site doesn’t seem to tell you what makes up the data.
http://www.theeconomicanalyst.com/content/stats-canada-discusses-household-debt though has very similiar looking graphs and it seems to be showing that the debt levels include mortgages. Note that the total mortgage is included, not an estimate of how much you’d owe the bank if you sold your house in today’s market.
That actually probably makes sense. According to http://www.statcan.gc.ca/pub/11-402-x/2011000/chap/fam/fam-eng.htm about 2/3 of Canadians own their own home. So 1/3 of the population would have no mortgage debt. As far back as 1971 the ownership rate was in excess of 70% so of the 2/3 of households that own their own home, there’s probably many that have little or no mortgage on it. When you average everyone together, you get a debt to income ratio to the turn of the 145% or so cited.
Does the term “disposable income” have a hard, scientific definition that is consistent across all nations and all the different partisan economists?
Or is it a completely meaningless term, entirely malleable, such as “global warming” ?
Yeah, it’s so tough to figure out what “global warming” could mean. Does it mean the temperature is going up or down? I just don’t know.
Well, if global warming was so coherent, then why did it transform without comment into the more generic “climate change”?
Because “global warming” became politicized.
Well, the globe is basically the only part that isn’t warming.
In Europe, climate change may increase the average temperature (as Europe is dependent on the jet stream and a large melt off of fresh water from Greenland on the top of the North Atlantic currents might flatten out the jet stream). Thus, if we say “global warming” we might assuming higher temperatures everywhere, which is unlikely to be the case. However, the net aggregate temperature of the earth is rising, so global warming is correct in that way, but might mislead you if you live in, say, soon to be colder Sweden.
Disposable income has a very specific definition in economics. I won’t do your homework for you, I’ll let google show you the answer.
In the old days when I paid for virtually everything by check or cash, I had zero debt. Now I pay for everything with plastic, but always pay the balances in full when they’re due.
So, if I pay off my credit cards every month (say $7000), am I considered to have $7000 in debt? Because I don’t.
It depends on how they’re measuring, but interpretations such as “your average balance” and “your credit limit” are not uncommon.
$3,500 is probably the best measure of your debt. Just because you can pay it off at the end of the month doesn’t mean it’s not debt. I can pay off my mortgage, but it’s still debt. You presumably aren’t net in debt.
“So, if I pay off my credit cards every month (say $7000), am I considered to have $7000 in debt? Because I don’t.”
It’s your Net debt that would matter. If you are paying off your credit cards at the end of the month, you probably have enough money in a checking account to cover your credit card bills at the end of the month. So the net debt would be close to 0, ignoring other debts, car notes, house, etc.
I don’t think so, but I could be wrong.
For example, people with a $600,000 house and a $200,000 mortgage are generally considered $200,000 in debt, even though they aren’t in net debt. It would surprise me, partially because it would be very hard to measure assets, if this analysis looked at net debt.
If you’re trying to get a loan, the lender won’t care (as much) what your average balance is; they’ll look at the minimum payment you would have to make on the maximum credit line you could draw. The rational being that they’re interested in what it would look like if things went badly.
On an indiidvual basis net debt makes more sense (although that needs to be tempered by liquidity). On the other hand, on a systemic level, maybe there’s something to be said for looking at gross household debt: the current net debt (or net worth) position is going to be affected by assets (esp real estate) that might be subject to a bubble; if the bubble implodes, the assets will be worth less but the debt will be the same debt. So gross household debt might be useful for looking at vulnerability to asset-price shocks.
The reason behind my question is that, if I pay zero interest (and zero card fees etc), how can it be considered debt at all? It should be considered as nil, not $7000 or $3500 or whatever. I understand that it’s a liability; but if all the “liabilities” of individuals or businesses who pay off their bills routinely without paying any interest are considered debt, that seems to be a lie. “Debt” that requires no servicing doesn’t really seem to be debt, at least as it pertains to the graph in question.
Maybe when people say “net debt” here, they’re really trying to get at “net worth,” which I don’t think is what this graph is showing. If I have asset-backed debt, like a mortgage, I’m still in debt, even if I might have positive net worth. Taken to an extreme, thinking about net debt, even at an individual level, can be very misleading when you’re talking about someone with a 97% CLTV mortgage, for example. Sure that’s positive when you net the asset against the liability, but it’s obviously risky. Gross debt matters.
In rk’s situation, I can’t think of any accounting system that wouldn’t consider him to be in a small amount of debt. He’s got principal due in the future – it requires servicing – someone else has a claim on his future cashflow and that’s the definition of debt. The interest rate as he describes it is irrelevant. Debt isn’t bad per se. What he’s doing doesn’t sound bad, and that doesn’t change if it makes him “in debt.”
I would say you have credit, not debt. But that’s common sense, not official jargon.
Common sense is that credit is the part of your card limit that you aren’t using, and debt is the part that you are using, isn’t it? Credit is something I can draw on, debt is something that I have to pay.
You are right. One problem is there don’t seem to be enough words to fully describe these concepts. At least I don’t know them. When you pay cash, in the 2 seconds between when you have the product and you hand them the cash, is that a credit and debt? No interest is charged when you pay your credit card bill immediately. The expectation is that payment is relatively immediate.
I read JWatts’ comment on ‘net debt’ to be referring to liquid assets (checking account) that could be applied against debt. In the credit card example, where you pay the $7,000 debt each month our of liquid assets, net debt is zero.
But you don’t get to net the value of your house against your mortgage.
In general you are correct that $7,000 per month in credit cards would show as debt. However, if you spending $7k on a card and paying off, the income must be ~$100K or so. Therefore you’re debt to income is 7%.
Yes, it moves the needle but not that much. (esp. compared to the US that will have the same thing.) It is mortgage debt that is the heavy hitter here.
The one reason is suspect Canada can have a higher debt:income ratio than the US is the incomes are much more equal and the middle class is a bigger portion of the income distribution. The middle class tends to have the worst debt/income ratios.
Perhaps not as bad as it looks:
Love the reasoning of that article: (1) some of the new debt is for cars, which is okay because cars are useful; (2) credit card balances are up 6%, but that’s okay because it’s at lower rates; (3) mortgage debt is up 7%, but that’s okay because it’s the lowest rate since mid-2009 (wow, lowest rate in three years!). This time really is different.
I don’t see what could possibly go wrong with that picture. After all, it’s not like interest rates could ever head upward. 😉
“Another difference between Canada and the USA is that you can’t get a 30 year fixed rate mortgage in Canada, but you can in the USA. Canadians have to use ARMs (adjustable rate mortgages), or a maximum of 10-year fixed rate, but most Canadians use 5-year fixed rate mortgages and hence have to refinance every 5 years.
The day the BoC start raising rates (which they will one day – see below chart), there will be homeowners in every market that cannot afford to refinance at higher rates and hence the default rate will rise.”
This is the key issue – if you laid out this chart as the ratio of interest payments to income, it would not look so dramatic. As long as rates stay low, things will be fine.
A significant increase in rates could spur a crisis – this is interest rate risk in a different form, not “too much debt” per se (and certainly not moralizing about “gluttony” like you see in the press)
Great news though – this is not an exogenous variable. The BoC can significantly change the interest rate picture by loosening or tightening monetary policy. They can also choose to allow some extra inflation, decreasing debt directly. And the BoC is very aware of this.
So the actual question is – will the Bank of Canada choose to blow up the economy? They hope not to have to make the choice, which is why they are working with other financial regulators to tighten lending requirements. We’ll see.
The BoC is prohibiting from allowing higher inflation by law.
If the graph excludes mortgage debt, it begins to make sense for Canadians.
For example, the latest Canadian disposable income stats show personal disposable income at $30,631 (Q1 2012 data). It doesn’t take much in Canada to get non-mortgage debt above that level. Large durable goods like cars can be significantly more expensive in Canada. Buy a decently equipped Ford Escape and you’ll be in the mid-$30,000 range. If you need a good truck (lots of trucks in Canada), you can easily wind up in the mid-40,000 range. Add in a personal loan for some home renovations and maybe a few thousand on credit cards, and bingo.
The thing is, the extremely low interest rate environment we’re in has caused a lot of people to take out personal loans to spend on home renovations (the government encourages this with rebates and tax credits), and people are stretching out amortizations on the theory that there’s no point in working hard to pay down a loan that’s only costing 3% annually in interest.
Whether this rate is unsustainable depends on other factors, like how much average mortgage equity Canadians have, how inflated the housing market is, how much government debt we have, etc. Canada’s per-capita share of government debt is now much lower than it is in the U.S. It ranges from about $39,000 in Ontario to a low of about $15,000 in Alberta. In the U.S. the per capita share of just the federal debt is about $45,000, and state debt is added on to that.
I’m not sure how to factor all that in to a ‘sustainability of debt’ index, but I don’t think it’s quite as bad as that chart indicates.
Comments on this entry are closed.
Previous post: Solve for the equilibrium
Next post: New edition of Ethnic Dining Guide
Email Tyler Cowen
Follow Tyler on Twitter
Email Alex Tabarrok
Follow Alex on Twitter
Subscribe in a reader
Follow Us on Twitter
Marginal Revolution on Twitter Counter.com
Get smart with the Thesis WordPress Theme from DIYthemes.