Commentary: the balance of payments and the savings gap.

Weale, Martin

Introduction

The recent national accounts showed the United Kingdom to be running a balance of payments deficit of 5.7 per cent of GDP in the third quarter of 2007, matching the record deficit incurred in 1988/9. In the period up to 1972, when exchange rates were fixed, a balance of payments deficit was a cause for concern, since there was the risk that it would not eventually be possible to meet the gap between imports and exports. The willingness of foreigners to invest in the United Kingdom might be limited and the foreign exchange reserves, which could be used to pay for imports, definitely were. With the change to floating exchange rates the exchange rate adjusts to clear the market. A balance of payments deficit does not then raise the same concerns, and this sometimes leads to the view that it is not very important. What should we make of the current balance of payments situation and what is it telling us about the state of the economy?

A balance of payments deficit is no longer a cause for concern per se, but it can be a symptom of economic imbalances indicating something wrong with the economy. These imbalances can be explored by understanding that the balance of payments deficit is equal to the gap between domestic investment and national saving. Thus a deficit arising because saving is low is one thing, while a deficit arising because of an expansion of investment opportunities is a different matter. Examination of the relationship between savings and investment in the United Kingdom allows us to form a preliminary view how far low saving rather than high investment is driving the deficit.

[FIGURE 1 OMITTED]

Figure 1 shows national savings and domestic investment measured as proportions of GDP. Both are measured gross of depreciation; nevertheless, the gap between them measures the balance of payments deficit. The broad picture is of savings being lower in the 1980s than earlier with a further clear decline in 1999 and another, which may prove less enduring, starting in 2005. The investment ratio is much where it was in the early 1960s, but we can see that there was a period of fairly high investment lasting from the mid-1960s until 1980. The late 1980s appear to have been a time of investment boom as well as rising consumption supported by falling saving, followed by low investment in the recession of the early 1990s and its aftermath.

Focusing now on the recent past, we can see that, compared with the period 2003-5, investment has risen by about 2 percentage points while saving has fallen by much the same amount. Thus a balance of payments deficit of below 2 per cent of GDP has been transformed into one close to 6 per cent of GDP.

The rise in investment has several possible explanations. Past work at the National Institute has drawn attention to the importance of variations in the risk premium as drivers of fluctuations in investment. But we have also noted the sharp rise in the UK labour force generated by flows of immigrants. We estimated in October 2006 that labour income had increased by 1.5 per cent in 2004-5 as a result of immigration. With a ratio of produced capital to GDP of 2.2 this points to an increase in the capital stock of 3.3 per cent of GDP as a result of immigration over this period. Immigration of course took place before 2004 and has continued in 2006 and 2007. Thus, while there are undoubtedly other factors present as well, the increase in investment can be linked to this. There is no obvious reason why this extra capital should be provided by domestic saving; it is perfectly reasonable for the balance of payments deficit to increase in response.

The fall in national saving of 2 percentage points is rather a different matter. If the country were plainly saving too much, one would be pleased by a fall in the savings ratio. But the evidence suggests (Khoman and Weale, 2006) that saving even in 2004 was a long way below the levels which would be required if each cohort were to pay its own way through its life and the situation has deteriorated since then. Thus, in this rather precise sense, the country was living well beyond its means in 2004 and is now living even further beyond them. The implication is inevitably that, while in the future the country's GDP may continue to perform well, the growth of its real income is likely to be rather less impressive. Since it is real income, rather than real GDP, which determines living standards, the low savings rate creates the prospect of a situation in which output growth disappoints because it does not deliver what people expect. There is probably some connection between the decline in saving and the fall in net property income from abroad (see pp. 41-2). However, even if this does recover as our forecast assumes, the basic savings shortfall remains a problem.

An alternative way of looking at the balance of payments deficit is to see it as the sum of the financial deficits of the Government and the private sector. We show these two deficits in figure 2. This makes clear that the Government's financial deficit has not changed very much, while the private sector's deficit has fallen by roughly the 4 percentage point worsening in the balance of payments position. Examination of the underlying data suggests that both saving and investment by the Government have been reasonably constant, and the dominant changes to both have been in the private sector.

We can identify three factors depressing private saving to various extents. The property boom has probably been the most important. This functions much like a budget deficit and has undoubtedly been an important factor discouraging saving and encouraging borrowing. Secondly, raising state benefits to old people is an obvious disincentive to save; the implications for retirement saving of the Government telling people that by raising benefits it has abolished pensioner poverty are clear. Thirdly, the general expectation that working lives are going to be longer than they were in the 1990s reduces the need for saving to some extent, although Barrell (2007) suggests that the effects of this are likely to be small.

[FIGURE 2 OMITTED]

Weaknesses of the Government's macroeconomic framework

The situation highlights the continuing weakness in the Government's macroeconomic framework. It has taken the view that the private sector's behaviour is none of its business except insofar as it affects the inflation rate. Thus the Government need worry only about the control of inflation--a task delegated to the Bank of England--and the state of the public finances. The flaw with this pre-Keynesian approach to the economy is now quite clear. A government cannot just brush aside the consequences of an inadequate savings rate, telling people that, having made decisions, they should live with the consequences. As is explained below, excess private borrowing creates pressures for public spending and a sensible policymaker would therefore be concerned about it as it happened instead of simply hoping for the best.

There are both short-term and long-term pressures. In the short term there can be pressure on governments to take steps to alleviate private sector debt. Thus Finland saw government borrowing rise by 8 per cent of GDP in the aftermath of its banking crisis in the early 1990s, while in the Far East in the late 1990s the fiscal impact of the financial crises was very much greater. It remains to be seen whether and how far bail-outs add to government debt in the United Kingdom or the United States.

In the United Kingdom the long-term pressures of low savings are very evident. There is considerable concern about the pensions crisis and the risk of poverty in old age, although few people manage to make the connection between this and aggregate saving. But the long-term consequence of low saving is that there is rising pressure to fund old age out of pay-as-you-go pension arrangements with obvious fiscal implications. Mrs Thatcher's Government attempted to address this problem by limiting the growth of the state pension to the rise in the retail price index and hoping that people would save more for retirement. But saving did not rise by enough to adjust for this. At the same time a strategy of telling people that they had to live with the consequences of their past saving did not prove to be politically sustainable. The result was that by the end of the twentieth century there were substantial increases in state benefits paid to old people. Despite its promise of benefit indexation in the short term, the current Government's strategy may turn out to be similar to Mrs Thatcher's. It is probably trying to limit growth of state benefits, but in the hope that saving in the new Personal Accounts does in fact lead to the required increase in private saving so that the limits on state benefits in the future can be made to stick.

The macroeconomic test of this will be to examine whether the ratio of national savings to national income rises once the scheme is introduced. If there is no evidence of a rise, then one will be able to conclude that Britain has not made any progress with solving its problem of structurally low saving.

The nature of the recent difficulty has been largely this. Our policy structure cannot handle the problem of domestic investment that is not very sensitive to interest rates combined with structurally low saving. In such circumstances interest rates may find themselves being used to keep consumption high on the grounds that consumption-led growth is better than no growth at all, a proposition more obvious in the short than the long run.

Thus a short-term solution to the current slow economic growth might be believed to be lower interest rates, in the hope that this depresses saving even further and raises optimism so that consumers keep spending. In other words the low interest rates which have been the source of our problems are also, mysteriously, the solution. But with saving already so low, the long-term difficulties that this gives rise to are inevitable. A more satisfactory outcome would be for growth to be maintained by higher domestic investment and a balance of payments improvement, so that saving would rise. The fact that the exchange rate has fallen fairly sharply since the summer points to some progress in remedying the structural imbalance of the economy. Indeed the fall may have come about because markets have realised the impossibility of sustaining a consumer/ property boom indefinitely, with the related expectation that future interest rates are going to be appreciably lower, compared with our neighbours, than was previously thought. This is discussed further on page 54.

Policies to influence saving

However, reliance on the exchange rate, which the Government does not control, to deliver a suitable level of saving would be a mistake. We therefore consider other policies which influence saving.

The fact that light-touch regulation of the banking sector has brought us to this present pass makes it extremely likely that the reviews of banking regulation currently underway will restrict banks' abilities to lend. This carries with it almost the automatic implication that in future borrowing will be restricted so that private saving is likely to be higher, certainly than it was in the last couple of years. Regulation is likely to add to the cost of banking, perhaps through raising the amounts of capital required or in other ways and this must raise lending rates relative to returns to savers.

In the current climate, it is hard to see political obstacles to tighter regulation. Other measures which might raise saving are, however, not so straightforward politically. One mechanism would be to impose a tax on consumer credit and mortgages; this could be done using the VAT mechanism and the requirement that creditors who wished to recover debts in court have to demonstrate that the tax has been paid would be enough to make it self-enforcing. The tax could be varied independently of the interest rate so as to regulate consumer spending. The proposal that banks and building societies should have substantial reserve requirements is very similar in outcome and perhaps politically easier to handle. It has the same effect of driving a wedge between the borrowing rate and the lending rate. Both approaches are logically more desirable than a return to credit rationing, since they give more choice and do not separate the population into two categories of people; those who can borrow fairly cheaply and those who cannot borrow at all.

A second, rather different route would be for the Government to set its fiscal position with reference to the overall level of borrowing in the economy. A tight fiscal position would mean that the Bank of England could set lower interest rates with taxes restraining the credit boom which would otherwise result. There is of course a view that private sector saving acts to offset changes in public borrowing so that such a change would have no impact. But the conditions needed for this to be the case are so stringent that they are most unlikely to be met in practice. A tight fiscal policy now would mean, in effect, that the Government is funding the payments it expects to have to make to old people in the future, i.e that it is saving on behalf of the private sector. In view of low private saving, it would not seem absurd for the Government eventually to run a surplus of 3 per cent of GDP instead of the current deficit of 3 per cent of GDP. Of course the private sector would react to that but in current circumstances it is unlikely that private saving could be reduced much further.

In the short term an obstacle to this is that it is a substantial change to the Government's existing position on fiscal policy. In the longer term, unless such an approach can be established with a cross-party consensus, voters are likely to reject governments which impose high taxes to run fiscal surpluses in order to raise saving. If people do not want to save for themselves, they will not rush, on election day, to thank a government which saves for them.

A third route might be to subsidise saving. Such a subsidy has to be paid for out of taxation. There would therefore be two effects. One would be an 'income' effect; the Government would be collecting taxes from people to add to their savings. This is not very different from a compulsory savings scheme, although the people who pay the taxes may not be those whose savings are subsidised. In presentational terms it would nevertheless look rather different from compulsory saving, at least if the tax increases required were not very obvious. Secondly, there is a substitution effect. The effective return is raised as a result of the subsidy. This may encourage extra savings, although probably not on a large scale.

Conclusions

The reality of Britain's savings gap is that all of these policies are probably needed to some extent. A willingness to run a structural surplus on the Government's budget, together with a non-regressive subsidy to saving, should probably be long-term features of the UK's economic landscape. A tightening of the conditions on which mortgages and consumer credit are available, using tools which are capable of short-term fine tuning, is likely to be a useful means of avoiding the sort of consumer boom which caused such large problems in the late 1980s and probably rather smaller problems in the past couple of years. But progress is unlikely to be made until policymakers understand the nature of the problem.

REFERENCES

Barrell, R. (2007), 'Retirement and pensions', National Institute Economic Review, 199, pp. 59-64.

Khoman, E. and Weale, M.R. (2006), 'The UK savings gap', National Institute Economic Review, pp. 97-III.