Friday, March 25, 2011

Timing Issues

According to Kevin Drum (h/t Coyote Blog), the states are the reason why the 2009 stimulus didn't work.
[A]t the same time the feds were spending more money, state governments were cutting back… [The Center on Budget and Policy Priorities has] data for all but six states, and on average for 2012, "those 44 states plan to spend 9.4 percent less than their states spent before the recession, adjusted for inflation." That's not just less than last year, it's less than 2008. That wiped out nearly the entire effect of the federal stimulus pacakge.
Kevin Drum seems to have an interesting definition of the phrase "at the same time." The stimulus bill was passed in February 2009, and most of the money was spent in 2009 and 2010. States' initial budget proposals for 2012, three years after the stimulus' passage, hardly seem relevant for the stimulus' failure. Even though some 5-10% of the stimulus will still be unspent by 2012, most of it has already been spent. If Drum wants to talk about what states were doing "at the same time" as the feds, why not use data on actual state spending for 2009 and 2010?

According to data from the BEA, expenditures by state and local governments did fall shortly before the stimulus was passed, from a peak of $2,218.5 billion in the third quarter of 2008 to just $2,171.5 in the first quarter of 2009. That's a whopping $47 billion drop, or just 6% of the $787 billion stimulus. Even assuming that state and local expenditures continued to grow past the peak in 2008-III at the same rate as population (inflation was -0.4% from July 2008 to December 2010), and summing across the two-and-a-half years since that peak, the difference between actual expenditures and "expected" expenditures is less than $200 billion. That's about a quarter of the size of the stimulus. And this is supposed to be the reason why the stimulus failed?

Sorry Kevin, the story just doesn't add up.

Sunday, March 20, 2011

Peak Debt

Friday's Financial Post ran an article by Paul Vieira on Canada's national debt, claiming that Harper has "wiped away" eleven years of debt progress. The old peak debt was $562.88 billion in 1997, and it is true that the current debt passed $562.88 billion on Friday. It now stands a little higher than that, and growing. So is Vieira right?

No, he isn't. He doesn't even try to give any context on the numbers he provides. Canada is in a much better fiscal position today than it was in 1997, and I've got the numbers and context to prove it.

Debt is often measured as a ratio to GDP. This is useful as it gives some perspective-- $500 billion is a lot more significant to the Canadian economy than it would be to the American economy. Thanks to economic growth, even the same level of debt is lower as a ratio of GDP. Nominal GDP in 1997 was about $883 billion, so $562.88 billion in 1997 dollars was 63.7% of GDP. But nominal GDP in 2010 was more than $1.6 trillion. A debt of $562.88 billion in 2010 dollars is only 34.7% of GDP-- only a little more than half what it was at the peak!

(According to the World Bank and StatCan, nominal GDP was $882,733 million in 1997, and $1,621,529 million in 2010. Why am I using nominal numbers rather than real? Because Vieira does, and I'll adjust for inflation further down.)

Another measure of debt is how much each citizen owes. In 1997, every person in Canada owed $18,771 of the $562.88 billion in national debt. Today, there are about 13.7% more people, so the same amount of national debt means that each person owes only $16,502. If the population continues to grow at the same rate (about 1% per year since 1997), and the debt continues to increase by $39 billion every year, we will reach the same amount of nominal-debt-per-person sometime in 2013-- which, again, doesn't even account for inflation.

(According to the World Bank, Canada's population in 1997 was 29,987,200, and grew to 33,739,900 by 2009-- 2010 is not yet available. StatCan's number for 2009 is slightly lower, at 33,720,200, but essentially the same, while the StatCan number for 2010 is 34,108,800.)

Five hundred billion today doesn't buy as much as five hundred billion in 1997 bought, thanks to inflation. To be more specific, you would need $725.55 billion in 2010 to buy what $562.88 billion bought in 1997-- a difference of $162.67 billion. Even if the Conservative government fails to reduce the deficit at all, and we keep adding $39 billion of debt every year, it will take more than four years to reach the previous peak debt.

(According to StatCan, CPI was at 90.4 in 1997, and 116.5 in 2010, with 2002=100. By setting the CPI in 1997 equal to 100, we get a value in 2010 of 128.9. This means prices in 2010 were 28.9% higher than they were in 1997.)

Putting It All Together
Adjusting for population growth and inflation, the national debt-per-person is just 68% what it was in 1997. As a percentage of GDP per capita, the national debt-per-person has fallen from 63.8% in 1997 to just 34.9%. In other words, in 1997 the average Canadian would have had to give up 63.8% of their income to pay their portion of the debt. In 2010, the average Canadian would only have to give up 34.9% of their income to pay their portion of the debt.

Make no mistake, I would love for Canada to return to a budget surplus, and the sooner the better. But misrepresenting the data is not the way to get there.