Talking about public debt: dumb, dumber and dumbest

As austerity is all the rage among policy making elites I thought it would be a good time to talk about ways of measuring public debt.  Here I will deal with the dumb way, the dumber way and dumbest way to talk about debt.

The usual way to talk about public debt is to express it as a ratio of debt to GDP.  The first problem wit this metric, however, is that there is no agreement between economists on what is a stable level of debt to GDP.  There does seem to be some agreement that sovereign governments (currency sovereigns) with their own nationally issued currency do not face any hard boundary on the debt to GDP ratio.  They may be output constrained, but they are never money (debt) constrained.  It is also generally agreed that subservient governments (those without central banks and their own currency, i.e., a province, a state, a member of the EMU) are money (debt) constrained.  Although, even here, there is no agreement on debt to GDP ratios.  the EU and IMF have chosen 120% as the health threshold for Greece.  None of the Canadian provinces are  even close to that level even if you throw in federal debt.  But then why would you throw in federal debt?  The federal government can always honour its Canadian dollar denominated debt obligations.

Lastly, when talking about GDP it can be measured quarterly, yearly or by the decade.  If I measured Canada’s debt to quarterly GDP I would, by definition, quadruple the ratio.  Similarly, if I measured debt as a percent of GDP per decade, I could reduce the ratio by more than a factor of ten (assuming the economy grows).

So which one do we choose?  If we think of GDP as an income stream it probably makes more sense to measure debt as ratio of GDP over 10 to 20 years (or maybe based on the term structure of the public debt).

Think about it.  When you go to bank for a mortgage they try to figure out what your income will likely be over the amortization period not one year, or one quarter.  So debt to annual GDP ratios are a pretty dumb way to talk about public debt.

Take heart though, there even dumber ways to talk about public debt.  Unfortunately for Newfoundland and Labrador, their premier has decided to have the dumber conversation.  The Globe reports that:

She [ms Dunderdale] also wants to bring down per capita debt – the highest in Canada – to the national average within 10 years.

I appreciate it is hard to sell soft-austerity to what was once a have-not province, that went without for a generation or more, only to be swamped by a tsunami of oil and gas profits.  Nonetheless, talking about debt per capita while not the dumbest conversation to have is dumber than it ought to be.

Measuring debt per capita is even more meaningless than debt per annual GDP.  Here is why.  First, demographic composition.  As Lana Payne, head of the Newfoundland and Labrador Federation of Labour, has pointed out, population increases can be driven by a growth in the economically inactive population.  Your debt per capita ratio goes down but your debt per economically active has gone-up.

Second, and somewhat linked to first, debt per capita says nothing about income.  What matters in Newfoundland (or anywhere else) is debt as a percent of GDP (subject to the limits discussed above).  Dunderdale is acting as though that when a family makes a mortgage application it is accepted or declined based on the size of the family and not household income.  This is disingenuous.  Ms Dunderdale knows better and so does her finance minister and so do the economists working for them.

The only dumber way, the dumbest way, is to talk about absolute public debt in nominal terms.

The real issue about public debt and budget deficits is what they are being used for and the capacity of the economy to absorb those investments.  The federal government, Ontario and Newfoundland and Labrador should be having that conversation.  Instead they are rehearsing respectable dogma.