How bond markets can learn to love public debt

Greetings Free Lunch readers.

In the meagre battle of ideas that the Labour Party leadership contest has thrown up, which I covered last week, one claim deserves a lot more attention than it has received. Lord Jim O’Neill, a former Treasury minister and Goldman Sachs chief economist, has suggested that bond markets may be quite happy to accommodate more government borrowing if it is used for spending they agree should boost growth. That is an interesting idea — because it should be a truism but is actually deeply controversial, and well beyond Labour and, indeed, the UK.

Every economic policy discussion in Europe today will in some way or other include the question of public investment — whether more is needed and, if so, whether it is possible.

I have long argued that the answer to the first is “yes”. Not only do we face important structural changes (towards digitalisation and decarbonisation) that involve whole new physical infrastructures, we also have years of under-investment to make up for. An FT deep dive into the money Germany belatedly has to spend to fix its long-neglected railways is a good illustration. We need to be running even just to stay still.

See the chart below. (I have struggled to get clean data for the UK — yet another casualty of Brexit is that Eurostat no longer gives nice, internationally comparable data for the UK. But for what it’s worth, the numbers I have worked with show the same dip in net investment in the 2010s.)

And this is not all. Rising geopolitical risk requires us to have better (but ideally dual-use) infrastructure for defence. Such threats, together with other sources of instability (such as climate change-induced extreme weather patterns), also imply more frequent and bigger supply shocks, which worsen the trade-off between inflation and economic output that central banks must navigate. Greater investment in resilient productive capacity can ameliorate it.

(Private investment matters too, of course. But a lot of infrastructure is a public good and needs some degree of public financing. And because public goods affect the profitability of private investments, the right kind of public investment boost can “crowd in” the private sector, while a public investment drought can hold it back from investing.)

Bond market investors know this too. That’s why I think O’Neill’s claim is (or should be) a truism: the right kind of borrowing for the right kind of investment is likely to make markets more willing to lend to governments.

But the standard set-up for monitoring public finances — institutions such as the Office for Budget Responsibility in the UK and the European Commission’s European Semester, as well as national fiscal watchdogs in the EU — is not conducive to this sort of thinking. While I’m obviously exaggerating for effect, they largely enforce various formulations of “higher debt-to-GDP ratios = bad”.

But as O’Neill also suggests, once you accept the economic logic, it shouldn’t be that hard to work within the rules, with minimal tweaks, to permit more borrowing and investments that markets actually welcome.

The first is to take more seriously the possibility that investments can improve growth. Most fiscal rules straightforwardly estimate the debt profiles caused by government borrowing choices, but are extremely reluctant to quantify the possible growth effects of the spending this pays for.

This is understandable: the watchdogs’ mandates and the avowed purpose of the rules they defend are to avoid excessive debt. It is natural to take a conservative stance against politicians’ obvious incentives to exaggerate the growth benefits and thereby avoid the hard choices. But it is also economically nonsensical to ignore what growth benefits there may be, let alone fail to discriminate between more and less promising projects, as far as we can know them.

To fix ideas, think about that huge potential white elephant that is Britain’s HS2 high-speed train project. It may now cost about £100bn. That amounts to about 3.5 per cent of 2025 UK GDP. But suppose we could be reasonably confident that it would raise the UK’s annual growth rate by 0.1 percentage point once operational (by allowing productive agglomeration effects along its stops, for example, and denser economic networks between some of the countries’ biggest cities). Then, over a period of about 40 years, the project would make the public debt-to-GDP ratio lower than otherwise even if funded completely by borrowing.

Governments, then, should do the work of knowing more. As O’Neill put it, they should identify spending with “high positive economic multipliers”.

Second, count assets as well as liabilities. The Labour government made one sensible change to the government debt rule: counting the debt net of financial assets. Increasing debt is no big threat to fiscal sustainability, after all, if it is matched by financial assets that reduce the net burden on the public finances.

But this holds, if not as strongly, for many non-financial assets too, at least those that can be sold or otherwise monetised in a pinch. So a second way to make fiscal rules more conducive to public sector prosperity is to make greater allowance for collateralised lending to finance infrastructure that can be sold or otherwise generate a cash return.

Public housing is the most obvious possibility. So is much transport and energy infrastructure. This could be kept separate from the regular fiscal balance sheet, for example through standalone development corporations with the authority to borrow to finance infrastructure. (The Tribune Group within Labour, which I discussed last week, has alluded to this.) There is a case, however, for giving such borrowing the backing of the government balance sheet. This is both to keep things honest and transparent, but also because a public guarantee will allow the standalone entities to borrow more cheaply (and markets may assume there is one anyway). But such guarantees should be accounted for net of the collateral — the things being built which could be handed to creditors in extremis. The result would be a much lower rise in the debt measure that actually makes economic sense and matters to intelligent investors.

Third, use longer time horizons. UK fiscal rules look at debt and deficits three years ahead. The EU’s rules look four to seven years ahead. Governments, however, often borrow for longer: the average remaining maturity of most EU countries’ debt is about eight years. And many bonds are issued for much longer maturities. Logically, the debt-to-GDP ratio that matters is the one for the year when the debt falls due. Fiscal sustainability metrics should reflect this. If that encourages more long-term borrowing — to match the long life of infrastructure investments — so much the better.

Other readables

● The FT sides with the Pope over Argentina’s president when it comes to AI.

● The UK’s productivity may be growing faster than most people think, Chris Giles writes.

● China has cut its oil imports almost in half, and nobody quite understands how.

● Yuan Yang, Labour MP and a former FT colleague, writes wisely on politicians and the bond market, calling for “the art of working within constraints . . . while, at the same time, expanding what is possible”.

● One of the things that has held Africa back is the relative lack of economic integration — too few economic ties and too little trade within the continent itself. When it comes to tourism, things may be changing.