A guest post by John Crompton on how the chancellor can turn the revelation of a £22 billion black hole in Britain's public finances to her and the country's advantage
Extraordinarily, the necessary work to switch to accrual accounting was initiated by Ken Clarke as chancellor more than 30 years ago. It then took another 20 years to get the systems in place. Yet even today the accounts are published two years late. There seems to be huge institutional resistance plus lack of political interest (for perhaps obvious reasons)...
Excellent account of what’s wrong with current public sector accounting, for pensions as well as capital assets. (Have comment on pensions which I’ll make separately.) Not clear to me what rules for fiscal sustainability you’d build on these excellent foundations.
In New Zealand, which adopted this approach more than 30 years ago and is still I think the only country to have done so, net worth is one of a set of fiscal principles (not rules) used to guide fiscal policymaking. The emphasis in NZ is on transparency in the public finances, which is impossible without proper accounting.
I recall something similar - resource accounting- being advocated at the beginning of the New Labour period, but it was put to one side by the introduction of the 40% to gdp sustainable investment rule, which constrained productive public investment below economically optimal levels as will Rachel Reeves debt to gdp rule.
Policy reform should focus on institutional reforms that will strengthen her fiscal sustainability credentials with market and other stakeholders, working in parallel with a strengthened Office of Budget Responsibility (OBR) remit to provide future scope and space to increase public investment to a level more supportive of the government’s overarching growth imperative.
Such institutional reforms to improve the quality and delivery of public investment must therefore be made integral, and not subsidiary, to the design and operation of fiscal rule reform. Both need to be inter-linked.
Long-term public investment decisions should be made demonstrably subject to a much robust and transparent selection, prioritisation, and delivery capacity appraisal process, capable of objectively ranking projects based on their clearly demonstrated economic and social returns relative to their claimed cost and capacity to deliver on time and on budget.
The remit of the well-established National Infrastructure Commission (NIC) should be provided with a statutory remit to assist each government department to publish an annual Departmental Investment Plan (DIP where projects are prioritised according to their estimated economic and social return. This would avoid the need for a new organisation and would work with the grain of the NIC’s existing institutional role and expertise.
It should also be freed from its current Treasury-set fiscal remit. Arbitrary limits on investment levels would undermine any effort of the NIC, given the lumpiness and interdependency of many projects, to select, rank, and sequence, projects efficiently.
1. I agree that issuing Gilts to a fund to meet existing and future pensions claims would, in a sense, be a "book-keeping" change. But this would have the very real effect of bringing these obligations inside the definition of public debt. This would make for better transparency, and would also remove the ability under current fiscal rules simply to ignore the very substantial pension costs of employing public sector workers. The effect would be roughly to double the amount of debt that is recognised by our fiscal rules framework, but at least we would know it was there, and government would be forced to take these liabilities into account in forming spending plans.
2. The "borrow and invest" strategy is more radical, but I believe it has merit if executed in a sensible way. This would involve borrowing at a steady rate, thought the cycle, and investing proceeds in a diversified, global portfolio. Over time, this strategy is extremely likely to yield returns significantly in excess of the UK's cost of borrowing (we assume 3%, though expectations would depend on the exact design of the portfolio),provided that we don't trash our credit - which is a working assumption for all government finance, and indeed, for the way that our economy works as a whole). It would also provide a large, liquid pool of mainly non-sterling assets - a useful thing to have on one's balance sheet in a crisis.
3. If we don't do this, current and future pension obligations will eventually translate into cash, and unless we in crease government revenues, this will mean more borrowing. This is one of the reasons why last year's OBR fiscal sustainability report projected an increase in government debt to 310% of GDP over a 50 year period. The advantage of our proposal is that while we would access the debt markets earlier, we would immediately start to capture investment returns. Once the pensions liabilities were fully funded, this would mean an improvement in government finances of 3% of GDP per year. This could be achieved inside a generation, in our view.
4. Finally, all this depends on the government having continued access to debt markets. But the steps we are proposing should provide just that, in a credible and convincing way, and should also materially improve revenue generation for the government, both through the liability management measures discussed here, and through improved asset management.
I completely agree about transparency. (Though I recall some evidence that the financial markets are irrationally less willing to lend to governments with funded pension schemes, and a purely nominal increase in announced debt would move the market.)
But government borrowing in order to create a sovereign wealth fund to invest in private sector assets does not appeal. Yes, it should achieve a return greater than the gilt rate; but because the UK government is a monopoly seller of gilts, marginal revenue is less than price, the gilt rate rises with borrowing volume and it’s the marginal rate on gilts that is relevant. Government borrowing also public investment, and unless the sovereign wealth fund can get a return higher than the test discount rate on public investment, the government should stick to its knitting,
John Crompton’s excellent piece states “Even more dramatic are the opportunities opened up by applying the net worth framework to public sector pensions. For a start, these would be treated as the debt obligations that they really are. If these pension obligations had been recognised as such when they occurred, and been funded by taxpayer contributions to an investment fund, as would be required in the private sector, then Britain’s public finances would be enormously better off now. Even if the government were to fund these pensions today, by borrowing and investing, it could expect a near-certain long-run return on investments well in excess of the cost of servicing debt. Over a generation, the obligations could be fully funded, and public finances improved by around 3% of GDP.”
I agree that clear public accounting for pension obligations should recognise that pension obligations are almost like gilts (only “almost” because governments have changed indexing rules on public pensions in a way that could not be done with gilts).
But I disagree that moving to fully-funded pensions would enormously improve the public finances. Consider the path to full funding. Step 1 would be to set up pension funds, and fund them fully with newly issued gilts to a value of, on current estimates, just over £2.5 trillion. This would be a dramatic step, but actually only a book-keeping change; making transparent what is currently opaque. Step 2 would follow logically, with new employee and employer pension contributions invested in gilts. Again, only a book-keeping change. UK public finances are not enormously better off – they are unchanged.
Step 3, however, could be to allow the public sector pension funds to diversify into equities and private sector bonds. On the face of it, this would not be a big positive step for the public finances. The Treasury would lose a captive flow of funds into gilts. More important, the Treasury is a monopoly seller of gilts, so the marginal return on gilts should be higher than the market rate (and indeed should be equal to the rate of return on public investment, if the borrowing level is optimised). Can the public sector really do better by investing in private financial assets than in real public assets?
Higher wages should lead to higher tax revenues. But this would continue to be captured in the budget forecasts. Investing in human capital is important but it doesn't create a public asset, in the same way that companies don't capitalise their investment in staff training. In fact this was one of the fiddles that Gordon Brown pursued with his (otherwise sensible) golden rule, with the result that we have the hopeless fiscal rules that we have today. Adopting proper accounting standards would make it much harder to blur the lines between spending and investment.
Most government spending will have a positive impact on UK tax revenues, the exact amount will vary somewhat, but this is very much a second order effect. Switching from capital expenditure to paying wages might increase the margin increase the tax take if this means reducing the amount of imported capital equipment, but I do not believe the Treasury does, or should, take this into account when switching spending fro one use to another.
I'm far from being an expert, but this sounds like common sense. So why wasn't this change made decades ago?
Extraordinarily, the necessary work to switch to accrual accounting was initiated by Ken Clarke as chancellor more than 30 years ago. It then took another 20 years to get the systems in place. Yet even today the accounts are published two years late. There seems to be huge institutional resistance plus lack of political interest (for perhaps obvious reasons)...
Excellent account of what’s wrong with current public sector accounting, for pensions as well as capital assets. (Have comment on pensions which I’ll make separately.) Not clear to me what rules for fiscal sustainability you’d build on these excellent foundations.
In New Zealand, which adopted this approach more than 30 years ago and is still I think the only country to have done so, net worth is one of a set of fiscal principles (not rules) used to guide fiscal policymaking. The emphasis in NZ is on transparency in the public finances, which is impossible without proper accounting.
Great to catch up again with your wisdom, John!
Brilliant exposition but unlikely that Reeves will change her fiscal policies.
I recall something similar - resource accounting- being advocated at the beginning of the New Labour period, but it was put to one side by the introduction of the 40% to gdp sustainable investment rule, which constrained productive public investment below economically optimal levels as will Rachel Reeves debt to gdp rule.
Policy reform should focus on institutional reforms that will strengthen her fiscal sustainability credentials with market and other stakeholders, working in parallel with a strengthened Office of Budget Responsibility (OBR) remit to provide future scope and space to increase public investment to a level more supportive of the government’s overarching growth imperative.
Such institutional reforms to improve the quality and delivery of public investment must therefore be made integral, and not subsidiary, to the design and operation of fiscal rule reform. Both need to be inter-linked.
Long-term public investment decisions should be made demonstrably subject to a much robust and transparent selection, prioritisation, and delivery capacity appraisal process, capable of objectively ranking projects based on their clearly demonstrated economic and social returns relative to their claimed cost and capacity to deliver on time and on budget.
The remit of the well-established National Infrastructure Commission (NIC) should be provided with a statutory remit to assist each government department to publish an annual Departmental Investment Plan (DIP where projects are prioritised according to their estimated economic and social return. This would avoid the need for a new organisation and would work with the grain of the NIC’s existing institutional role and expertise.
It should also be freed from its current Treasury-set fiscal remit. Arbitrary limits on investment levels would undermine any effort of the NIC, given the lumpiness and interdependency of many projects, to select, rank, and sequence, projects efficiently.
Many thanks. A few thoughts:
1. I agree that issuing Gilts to a fund to meet existing and future pensions claims would, in a sense, be a "book-keeping" change. But this would have the very real effect of bringing these obligations inside the definition of public debt. This would make for better transparency, and would also remove the ability under current fiscal rules simply to ignore the very substantial pension costs of employing public sector workers. The effect would be roughly to double the amount of debt that is recognised by our fiscal rules framework, but at least we would know it was there, and government would be forced to take these liabilities into account in forming spending plans.
2. The "borrow and invest" strategy is more radical, but I believe it has merit if executed in a sensible way. This would involve borrowing at a steady rate, thought the cycle, and investing proceeds in a diversified, global portfolio. Over time, this strategy is extremely likely to yield returns significantly in excess of the UK's cost of borrowing (we assume 3%, though expectations would depend on the exact design of the portfolio),provided that we don't trash our credit - which is a working assumption for all government finance, and indeed, for the way that our economy works as a whole). It would also provide a large, liquid pool of mainly non-sterling assets - a useful thing to have on one's balance sheet in a crisis.
3. If we don't do this, current and future pension obligations will eventually translate into cash, and unless we in crease government revenues, this will mean more borrowing. This is one of the reasons why last year's OBR fiscal sustainability report projected an increase in government debt to 310% of GDP over a 50 year period. The advantage of our proposal is that while we would access the debt markets earlier, we would immediately start to capture investment returns. Once the pensions liabilities were fully funded, this would mean an improvement in government finances of 3% of GDP per year. This could be achieved inside a generation, in our view.
4. Finally, all this depends on the government having continued access to debt markets. But the steps we are proposing should provide just that, in a credible and convincing way, and should also materially improve revenue generation for the government, both through the liability management measures discussed here, and through improved asset management.
I completely agree about transparency. (Though I recall some evidence that the financial markets are irrationally less willing to lend to governments with funded pension schemes, and a purely nominal increase in announced debt would move the market.)
But government borrowing in order to create a sovereign wealth fund to invest in private sector assets does not appeal. Yes, it should achieve a return greater than the gilt rate; but because the UK government is a monopoly seller of gilts, marginal revenue is less than price, the gilt rate rises with borrowing volume and it’s the marginal rate on gilts that is relevant. Government borrowing also public investment, and unless the sovereign wealth fund can get a return higher than the test discount rate on public investment, the government should stick to its knitting,
John Crompton’s excellent piece states “Even more dramatic are the opportunities opened up by applying the net worth framework to public sector pensions. For a start, these would be treated as the debt obligations that they really are. If these pension obligations had been recognised as such when they occurred, and been funded by taxpayer contributions to an investment fund, as would be required in the private sector, then Britain’s public finances would be enormously better off now. Even if the government were to fund these pensions today, by borrowing and investing, it could expect a near-certain long-run return on investments well in excess of the cost of servicing debt. Over a generation, the obligations could be fully funded, and public finances improved by around 3% of GDP.”
I agree that clear public accounting for pension obligations should recognise that pension obligations are almost like gilts (only “almost” because governments have changed indexing rules on public pensions in a way that could not be done with gilts).
But I disagree that moving to fully-funded pensions would enormously improve the public finances. Consider the path to full funding. Step 1 would be to set up pension funds, and fund them fully with newly issued gilts to a value of, on current estimates, just over £2.5 trillion. This would be a dramatic step, but actually only a book-keeping change; making transparent what is currently opaque. Step 2 would follow logically, with new employee and employer pension contributions invested in gilts. Again, only a book-keeping change. UK public finances are not enormously better off – they are unchanged.
Step 3, however, could be to allow the public sector pension funds to diversify into equities and private sector bonds. On the face of it, this would not be a big positive step for the public finances. The Treasury would lose a captive flow of funds into gilts. More important, the Treasury is a monopoly seller of gilts, so the marginal return on gilts should be higher than the market rate (and indeed should be equal to the rate of return on public investment, if the borrowing level is optimised). Can the public sector really do better by investing in private financial assets than in real public assets?
Please see my comment above, sorry I wrote in the wrong box!
In one sense I think of them in the same way as capital investment.
Higher wages should lead to higher tax revenues. But this would continue to be captured in the budget forecasts. Investing in human capital is important but it doesn't create a public asset, in the same way that companies don't capitalise their investment in staff training. In fact this was one of the fiddles that Gordon Brown pursued with his (otherwise sensible) golden rule, with the result that we have the hopeless fiscal rules that we have today. Adopting proper accounting standards would make it much harder to blur the lines between spending and investment.
I'm not even close to being savvy on this topic. Just asking about tax revenues effect of investing in our NHS people, for example.
Doesn't the £9.4bn paid out for wages lead to significant tax revenues?
Most government spending will have a positive impact on UK tax revenues, the exact amount will vary somewhat, but this is very much a second order effect. Switching from capital expenditure to paying wages might increase the margin increase the tax take if this means reducing the amount of imported capital equipment, but I do not believe the Treasury does, or should, take this into account when switching spending fro one use to another.