Some Canadian conservatives were predictably angry last week when the Federal Liberal government highlighted Canada’s relatively low net debt. They want to instead focus on the gross debt, which moved in a different direction. (The figure above shows the gross and net financial liabilities of the “general government” — includes provincial governments. Canadian provincial governments have a big economic footprint, and in order for the debt figures to be comparable to international peers, you want to look at the general government. Otherwise, the Canadian government looks to be a misleadingly small part of the economy.)
I do not have time to dig into all of the sources of the divergences between gross debt and net financial liabilities, but I want to discuss one that was raised — the role of public sector pension funds. My old employer was the Quebec public asset manager that has a long history, and the Canadian Federal government created its own in 1997, and has grown to a very large size (by the standards of the Canadian market).
The conservative argument was along the lines that it was unfair to subtract those pension assets, since they are earmarked for a specific purpose, and thus do not truly offset government debt. If one wants to model the debt dynamics and things like interest rate sensitivity, we do need to keep in mind the gross debt. However, if one is worried about “sustainability,” it is entirely on brand for free marketeers to complain about bloated gross debt levels since that is an entirely predictable outcome of the creation of those plans — which were favoured by earlier generations of free marketeers (or even the people complaining).
I am going to put aside provincial plans herein. Provinces are sub-sovereigns, and face default risk. They need to run their pension plans on similar principles to private pension plans, although they have demographic advantages versus the private sector — populations are growing, whereas most companies with defined benefit plans have shrinking workforces.
There are three defensible ways of running the Federal pension plans (note that there are plans for Federal employees, as well as the universal one that offers minimal payments).
Pure pay-as-you-go (PAYGO): take in pension premiums as general revenue, and pay out benefits as a general expense. Nothing ends up on the balance sheet. (Contingent claims — projected payments — are not on the balance sheet.) No effect on gross or net debt.
Pay-as-you-go with a fictitious trust fund: pension premiums are used to buy non-negotiable Government of Canada bonds. This has exactly the same economic effect as (1) — premiums go into general revenue, payments are a general expense, but we have fictitious bonds on the government’s balance sheet. The fictitious bonds will end up in “gross debt,” but since they are also assets owned by the consolidated government, they get subtracted to arrive at a lower net debt.
Pension premiums are funnelled into a government-controlled asset manager, which buys financial assets from the private sector (including buying Federal government bonds in the secondary market). Since the pension premiums are no longer going into government general revenue, this is equivalent to a lower tax rate, and so cash deficits are larger and the government needs to issue more negotiable bonds. (Note that this is not the accounting deficit.) The exact economic effect of this change is uncertain (as I discuss below), but the simplest to understand outcome is that this option is equivalent to the Federal government issuing debt to buy financial assets. This is doing exactly what finance professors love: levering up the government’s balance sheet to “arbitrage” the difference between the risk-free rate and the returns on a diversified financial portfolio.
If we look at the difference between (1) and (2), we see that the cash flows are the same, all that happens is that there are fictitious government bonds showing up in (2). So long as we assume that the population-at-large is not as gullible as mainstream economics professors, the amounts of fictitious bonds in existence will not affect behaviour. This means that “pension contributions” are fungible with “taxes”: if we change the “pension contribution,” we need to make an offsetting change to taxes to get the same economic outcome.1
Option (3) adds a wrinkle: the government is issuing debt to buy private securities. On paper, this can be accomplished purely by portfolio reallocations, and have no income effects on the economy. However, private sector asset prices will presumably rise, and private debt issuance might pick up. As a result, one might expect that there is a small positive effect on economic growth — but I would expect the multiplier be to be pretty close to zero. Offsetting this growth-enhancing effect is the reality that the governmental pension fund is a cash flow sink for dividends and interest payments from the private economy. This is economically equivalent to a tax — which implies a need for looser fiscal policy to keep the same level of economic output.
In any event, it is straightforward that “running government pensions like a private pension” — a policy favoured by neoliberals — is going to result in a larger governmental balance sheet. Complaining about greater gross debt by the same neoliberals is akin to someone who murdered their parents asking for mercy since they were newly orphaned.
Let us say that we want to increase “pension contributions” by $100 million but keep cash flows in the economy the same. We relabel $100 million in “taxes” as “pension contributions” and add $100 million in fictious bonds to the government’s balance sheet.
Interesting to see the difference in the 'underlying trend' in your leading figure pre- and post-2008: pre-2008 gross debt tracked well with net financial liability; post-2008 the structural effect is of a divergence between gross debt and net financial liabilities. I assume that this is a result of central bank ZIRP policy where central bank balance sheets were expanded but government debt could be rolled over at lower interest rates?
There is a fourth way - just pay pensions, ie, with no premiums whatsoever. As we all know, by now, Federal government cheques can never bounce. So just have the Canadian government spend money directly into retirees bank accounts and have done with it. Most of the money will be immediately recycled back into the economy and what isn't can be taxed back. Premiums (and all other pay cheque deductions) are simply payroll taxes - the most regressive taxes imaginable.
The problem with option 3 - pension funds - is that they're compelled to take risks to make money to pay pensions. We saw the consequences after Liz Truss's budget in the UK - illiquid pension funds having to be shored up by the Central Bank. It makes a mockery of the whole scheme. If the state has to bail out pension funds, why not just have the state print pension cheques in the first place?
One might argue that pension funds invest in the stock market but if they are speculative rather than productive and if they, too, are held aloft by Central Bank's quantitative easing, you again come full circle to just having the Federal government directly providing pensions. And a wholly beneficial effect would be to reduce secondary markets, which would go a long way to stabilizing the financial system. Mind you, it would also eliminate a lot of well paid but essentially pointless fund manager jobs, which is why is quite unlikely to happen.