The Waiting Until We Can Afford It Argument - Part II

This more of an argument you hear for a one-time expense such as building new schools or buying new military aircraft, that goes something along the lines of:

"We already have a deficit, so this policy will just increase the deficit and we'll pay even more in interest.  But if we wait until we have a surplus, we can just use the some of the surplus money and we avoid making interest payments."

It's a flawed argument, because it fails to realize that there is always a cost of capital.

For a small business, the cost of capital is significantly higher when financed through debt.  If our business wants to borrow money to finance a project, we have to do so through a bank where we'll pay near usurious rates of interest (typically around 20%).  However, if we are debt-free and finance the project through cash-on-hand, our capital costs include the interest we lose by not investing in something else, such as a GIC.  Given that a 1-year GIC is paying less than 2% interest, the cost of capital is quite low.  As such, there will be a host of projects that make financial sense when financed through cash, but do not make sense when financed through debt. 

So why doesn't this hold for governments?

Roughly speaking, when the government runs a deficit and the debt increases, they finance that debt through a bond issue.  As such, the (nominal) cost of capital is the interest rate on that bond.

Over time, bonds mature and the government pays their face value.   To pay the face value, the government can either use cash or they can issue another bond.

So what happens when the government runs a surplus? Bonds mature, are paid out in cash from the surplus and no new bond needs to be issued.

If we finance a project during a time of surplus, then we reduce the amount of surplus that can be used to pay out maturing bonds, so less bonds can be paid out this way.  The remaining bonds need to be re-issued.  As such, the cost of capital for financing the project is the interest rate on that bond.

In both the cases of financing a project during a deficit and financing a project during a surplus, the cost of capital is the interest rate on the government bond.  That does not imply that the cost is the same - the interest rate on a bond can be different between a deficit period and a period of surplus.  Since deficits are associated with times of recession, and in a recession the Bank of Canada tends to lower the overnight rate, the nominal cost of capital, if anything, is likely to be lower during a deficit period than during a surplus.

The cost of capital is important, no matter if we are in a surplus or deficit position.  Waiting for a surplus does not make the cost of capital go away.

Comments

[Georgian]

How do you quantify investor confidence in a country with a balanced budget?

[Phil Koop]

Your basic point is obvious, but I have to question your accounting. A budget that is balanced is by definition financing all expenditures, including interest, out of current income. A surplus is money above and beyond this. So the opportunity cost of spending a surplus cannot be the refinancing of existing debt. It can be: 1. a foregone reduction in taxes, or 2. a foregone opportunity to call existing debt.

Not much government debt is callable (short of special legislative measures.) Debt that was callable on issuance would embed the option premium, negating your point about cheap money during recessions. A tax cut during a boom would goose the economy even more; not necessarily a good thing. If the cut were not sustainable during the next downturn, it would be procyclical, an automatic destabilizer. The ideal would seem to be to never run a recurring surplus, but to run persistant deficits such that the through-the-cycle rate of debt accumulation does not exceed the growth in the economy. Fine judgment is required! :-)

[Mike Moffatt]

Hi Phil,

It isn't so much that the new debt is getting recalled, just that it isn't being re-issued when it matures.

[Patrick]

"How do you quantify investor confidence in a country with a balanced budget?"

Balanced over what time frame? 1 nanosecond? 1 year? 10 years?

[Georgian]

Balanced over what time frame? 1 nanosecond? 1 year? 10 years?

Google: "balanced budget investor confidence". You'll find many examples over many time frames - so obviously it is not an abstract issue - but neglected here.

[Phil Koop]

Oh, do I just not know what "balanced budget" means? Apologies for that, and thanks for the education. (I had supposed that a balanced budget is one that pays debt as it matures, rather than rolling it over in perpetuity.)

[Patrick]

Georgian - Huh? Are we talking about the same thing?

'Investor confidence' is pretty easily observed in bond yields etc, so why borrow trouble? As Mike points out, the cost of capital is usually lower during recessions, and tax revenue higher during booms. Suggests to me that balancing the gov'ts budget over the business cycle would be a good starting point, whereas picking some arbitrary fixed time interval - like one year - would be extraordinarily stupid. Just look at the pain in US states with annual balanced budget requirements.

[Determinant]

Government debt may not be callable as such but it is most definitely for sale on the open market at predictable prices. There isn't much difference when the government wants to retire debt if they call it, purchase it or just plain redeem it without refinancing it. Around 2004 when we were running surpluses there were papers on Finance's website on this. The powers-that-be thought it was wiser to purchase smaller and "off-the"run" issues maturing in three years or less and concentrate the market on larger, more tradeable issues rather than just retiring each and every bond issue as it came due. The goal was to keep prices high and interest costs low.

[Erin Weir]

Mike, this mistitled post makes a correct argument about public financing of one-time investments. But it does not have much to do with CIT cuts and does not even mention them (outside of the title).

[Mike Moffatt]

Erin: Yeah, the title isn't the greatest. Just called it 'Part II' because it's a continuation of my thoughts from yesterday. Probably should call it something else.

[Mike Moffatt]

There.. fixed.

[Georgian]

Patrick,

If lowering CIT allegedly results in marginal investments, then running deficits (in fact extended in length due to CIT cuts) should have the opposite effect. No? Why not?

[RSJ]

A "balanced budget" is one in which debt is neither increasing nor decreasing. If you are not rolling over debt, then you are running surpluses, not balancing the budget.

But as GDP is rising, and you really care about debt/GDP, then you would never want a balanced budget, you would want a budget that in general is in deficit.

Moreover, as the weighted average risk-free rates are in general lower than the growth rate of GDP, you would need to run larger deficits.

For example, using historical U.S. data, if you assume an income growth rate of 3%, risk-free rates for maturities in the 6 YR range of 1.7%, and an "ideal" debt/GDP ratio of 60%, then you would need to run a *primary* deficit of about 1% of GDP every year to keep the debt/GDP level constant. This means a structural budget deficit of about 2% of GDP, across the business cycle.