From the Archives: Not What, When - Misunderestimating the Debt

Note: I'm going through and adding labels and fixing grammatical errors in my old posts (until I get bored and move on to another project). I figure while I'm at it, I should do some spring cleaning (ok, I'm a little behind schedule) on some of my draft posts that I never finished - what follows is the oldest draft post that remains in the draft folder, it was originally drafted on November 17, 2004 (any comments that I add today as I post will be marked with (ed: ...).
----


Lots of comments these days about what should be done with the Federal Government's surplus. Now, there are a number of arguments that can be made about whether the money should go to debt reduction or not. Arguments about intergenerational fairness, preparing for future economic shocks/rate changes, the impact of the debt on the economy/interest rates and a whole bunch more.

But for this post I'm going to ignore all that and just focus on one question: Which route will allow us to spend more money (be it on health care, tax cuts, child care, the military or whatever):

Option 1) Spending the surplus this year, and every year that follows or
Option 2) Applying the surplus to the debt this year and then spending it next year and every year that follows

To keep things simple, I'm going to assume that every year will be like this one: a $9 billion dollar surplus and an effective interest rate on federal debt of (roughly) 5%.

Under option 1, the amount we have to spend will be $9 billion this year, $9 billion next year, $9 billion the year after that and so on. Simple.

Under option 2, we spend $0 this year, $9.45 billion next year (because we are no longer paying 5% interest on $9 billion we payed off in year 1: $9 billion x 5% = $450 million), and then $9.45 billion the next year and so on.

Clearly, over time we will end up spending more money under option 2 (after 20 years) but to make a proper comparison, we need to consider the time value of money -i.e. that money spent now is worth more than money spent in the future.

If $1.00 spent this year is worth more to us than $1.05 spent next year, than we can get more value by spending the surplus now. So the government should only spend on those items which will gain a greater than 5% return (in revenue to the government)**.

My own opinion is that few options for government spending or tax cuts will yield the same return as paying down debt, a return which has the benefit of being nearly guaranteed.

But that's a minority opinion. Here's an article in the Toronto Star:
http://tinyurl.com/5mp4e (registration required)

If you don't feel like reading it, I'll save you the trouble (ed: especially since that page is long gone, and you don't need to register to look at the Star's website anymore!). Basically, anyone you can think of with any involvement in politics in Canada thinks that the money should go their cause/constituency/whatever and nobody thinks the full amount should go to debt reduction. Quelle surprise.



Why does this happen?

I can think of three reasons:

1) People don't really understand debt or interest
2) People are impatient - i.e. they'd rather have a $10 bill today then a $20 tomorrow (this just means they apply a high discount rate to their consumption).
3) Collective action problems - the various groups figure that if they don't make their case heard the government will give the money to the other side. They could get around this by framing their comments as 'The government should use this money for debt reduction but if it doesn't then it should go to our cause' but maybe that is too long/complicated a message for the sound-bite media.

Solutions:

(ed: this is as far as I got before I left off, back in 2004. As it turns out, maybe there was no foolproof solution...in retrospect the best solution might have been 'don't elect a conservative government', but then that's sort of a default solution to a lot of political problems, isn't it?)

---
** Technically, the required rate of return on any debt financed government spending should really be high enough to cover the cost of borrowing for that specific purpose (almost certainly higher than 5%). i.e. if the government is borrowing
to start (or finance) an experimental high tech firm, we would want a better than 5% rate of return to justify the investment because of the increased risk.