## Spencer Dale: Separating fact from fiction - household balance sheets and the economic outlook
Speech by Mr Spencer Dale, Executive Director and Chief Economist of the Bank of England, at a lunch hosted by the Exeter Chamber of Commerce, Exeter, 24 September 2009.
I would like to thank Andrew Benito, Rohan Churm and Richard Williams for their considerable help in preparing these remarks. The views expressed are my own and do not necessarily reflect those of other members of the Monetary Policy Committee.
* * *
I would like to thank the Exeter Chamber of Commerce for this invitation to speak to you today.
I have strong ties to this part of the country. My family moved to Exmouth when I was 16. My father still lives in Exmouth and my parents-in-law in Exeter. So I visit here often.
Exeter was the place that I first studied economics. My family moved to Devon just after I had completed my 'O levels' and I studied for my 'A levels' at Exeter college, just a few minutes walk from here. In fact after today's lunch, I am going back to Exeter college to see my old economics tutor - who still teaches 'A level' economics - and to spend some time meeting the current students.
So it is a real pleasure to be here.
The economy appears to have turned. Following the extraordinary falls in output at the end of last year and the beginning of this year, output appears to have stabilised. And the most recent output data and business surveys suggest that we are likely to see positive growth in the second half of 2009.
These indicators have prompted numerous headlines declaring an end to the recession. In a strict, technical sense that may be right - we may be moving into a period in which positive growth is resumed. But I would caution against drawing too much comfort from those headlines. Even if output does expand in the second half of this year, for many families and businesses it may still feel like we are in the economic doldrums. Unemployment is likely to continue rising for a period and firms will still face sluggish demand. The recovery may be slow and protracted.
This caution is often characterised as the Bank being gloomy. But I fear that misses the point. Let me be clear, compared with the situation we faced only six or nine months ago, the economic outlook has improved significantly. Output is no longer in free fall, the banking system has stabilised, confidence levels have improved, and the risk of a really bad economic depression has receded. To repeat, the economy appears to have turned an important corner on the road to recovery.
But we are only just starting along that road. The level of national output is estimated to have fallen by around 5.5% over the past year - the largest four-quarter fall experienced in the post-war period. Employment has also fallen substantially. These falls have resulted in a significant degree of economic slack opening up in the economy. Firms are operating below normal capacity; increasing numbers of people are unable to find jobs. This slack detracts from our well-being. And, if it persisted, would pull inflation down well below 2% - the target that I and my colleagues on the Monetary Policy Committee are mandated to aim for. To meet our target, we need to see a sustained period of robust growth that brings the level of activity better into balance with the supply capacity of our economy. It is a recovery in the levels of output and employment that matters, not just a return to positive growth.
BIS Review 114/2009
The turnaround in growth is underpinned by the extraordinary loosening in monetary policy undertaken by the Monetary Policy Committee over the past 12 months. Only a year ago, Bank Rate was still at 5% and quantitative easing was something we had studied only from afar. Since then, Bank Rate has been cut to an historical low of 0.5% and the Bank is a significant way through a programme of asset purchases totalling £175 billion. Economic activity is being supported further by the easing in fiscal policy, the substantial depreciation in sterling, and by the fact that the inventory adjustment appears to be beginning to run its course.
But there are a number of headwinds that are likely to slow the pace of recovery. Credit conditions remain tight as banks repair their balance sheets. And the global nature of the economic downturn further highlighted the need for a rebalancing of trade flows between countries with large current account surpluses and deficits. Both adjustments are likely to occur only gradually and each has the potential to derail the recovery.
I want to spend a little time today considering a third factor that will shape the recovery: the extent to which households need to rebuild their balance sheets following the big increase in household debt.
## Spend, spend, spend?
Much has been said about the increase in household debt, both in terms of its role in the build up to the financial crisis and the extent to which it may dampen the recovery. The theme of much of this comment is that our current economic difficulties are a payback for past excesses. Households, it is said, were seduced by easy credit and rising asset prices, leading to a debt fuelled spending boom. As a result, we now face a period of retrenchment as we make amends for our former spendthrift ways.
Like much conventional wisdom, these views contain a kernel of truth. But their repetition has led to exaggeration and distortion. Separating fact from fiction is vital when forming a view on the economic outlook. Consumer spending accounts for two-thirds of total demand in our economy; decisions made by consumers up and down the country will be critical in determining the strength and sustainability of the recovery.
A natural place to start would be with the 'consumption boom', which is often presumed to have preceded the bust. But in fact there was no such boom. Between the turn of this century and the onset of the financial crisis, consumer spending increased no faster than it had done, on average, over the past 40 years. 1 Moreover, the fraction of our national output consumed by households did not increase.
The kernel of truth in the consumption boom story is that changes in the distribution of our national income meant that this stable growth of aggregate consumer spending was associated with a sharp decline in the household savings ratio. The proportion of income that households set aside fell from close to 10% in 1997 to a little over 2% ten years later. Although there was no spending boom, households were making less provision for the future.
## Where did all the debt go?
But how does the absence of a spending boom fit with the headlines about the rise in debt. The big increase in household debt over the past decade is certainly not a myth. Household
1 Consumption growth was, however, unusually strong in the period 1996-2000. My colleague Andrew Sentance has stressed the importance of analysing trends in consumption over longer periods (eg Sentance 2007).
BIS Review 114/2009
debt as a proportion of income increased from 100% to 165% in the 10 years to 2007. If this big run up in debt was not used to finance a surge in spending, where did it all go?
The answer to this question lies in developments within the housing market. House prices trebled in the ten years to 2007. And mortgage debts were accumulated to pay for the housing that had become so much more expensive. The conventional wisdom that the sharp increase in household debt was associated with the house price boom of the past decade is well founded.
But what is less often appreciated is that much of that rise in household debt was matched by a comparable increase in the value of financial assets held by households. The value of housing assets purchased by households did increase, through additions and improvements to the housing stock. But the main counterpart to the rise in borrowing was increased saving. To make this point more concretely: while households accumulated an additional £1trillion of debt between 2000 and 2008, they also acquired over £750 billion of financial assets over the same period. 2
What was going on? To caricature things only a little, older households (or rather those trading down within the housing market) benefited from selling homes at much higher values. They were the winners. The big losers were younger households (or those trading up), paying more for those houses and accumulating higher debts. The money borrowed by young families ended up in the bank accounts of older households. That money wasn't used to fuel a spending boom. Rather, it was left on deposit or used to invest in other financial assets. The huge increase in borrowing by one part of the population was broadly matched by a big increase in saving by another part.
The increase in house prices over the decade to 2007 - and the massive financial flows associated with that appreciation - represent a huge redistribution of wealth between different households within our society.
Two important points should be drawn from this episode.
First, any analysis of the increase in household debt over the past decade has to pay equal attention to the record accumulation of financial assets. They are two sides of the same coin.
Second, changes in house prices do not result in significant changes in aggregate household wealth: for every winner gaining from higher house prices is a loser facing less affordable housing. But that does not mean that changes in house prices can not have significant implications for the macroeconomy. House price changes may directly affect aggregate spending both through their effect on the value of housing collateral, and hence on the ability of households' to access credit, and through the different propensity of winners and losers to spend their 'windfalls'.
## A debt hangover?
Even though the accumulation of debt was not driven by excessive consumption spending, the sharp falls in asset prices associated with the financial crisis have led to a marked deterioration in households' financial position - assets are worth less but the debts still need to be repaid. The market value of households' financial and housing assets has fallen by over 10% over the past two years. Does this mean that there is some form of hole in households' balance sheets that will weigh on the recovery as they seek to repair it?
2 This increase reflects purchases of assets. It does not include the returns from these investments.
Any real or perceived need for households to rebuild their balance sheets may play an active role in determining their levels of spending. 3 For example, the cost of financing debt and the uncertainty about future employment prospects may make households reluctant to exceed a certain level of net borrowing. In this case, the deterioration in the financial position of some households may cause them to spend less in order to repay debt or increase savings. The 'payback period' I mentioned earlier.
This possibility has focussed much recent attention on the state of households' balance sheets. But standard measures of households' financial position - such as the ratio of debt to assets - may exaggerate the severity of the problem. That is because these conventional measures suffer from a 'missing' asset and a 'missing' liability.
## Let me explain.
The 'missing' liability is the cost of purchasing a house in the future. This argument is directly related to the redistribution argument I discussed earlier. Measured in conventional terms, the marked decline in house prices seen over the past two years has led to a pronounced deterioration in households' balance sheets: house prices have declined by around 15% but the value of outstanding mortgage debt is largely unaffected. The lower level of house prices will eventually be reflected in a lower level of mortgage debt than otherwise as first-time buyers (and those trading up the housing ladder) enjoy the benefits of more affordable housing. But this effect will not fully materialise until the entire housing stock is turned over at the new lower price level, which may take a considerable time. In the meantime, not taking account of this missing liability exaggerates the deterioration in households' underlying position.
Moreover, by considering only the fall in prices of housing and financial assets, we could be exaggerating the extent to which people are actually less wealthy. For some households, these are their only form of assets. This is most obviously the case for the retired. But the largest asset owned by most people is not measured at all. It is the value of what they expect to earn in the future or, what economists refer to as 'human capital'. Most estimates suggest the value of future earnings is significantly greater than the sum of net financial wealth and gross housing assets which typically forms the basis for balance sheet measures. 4 The significance of this missing asset is that, even though recent events may have impaired households' future income prospects, the scale of this reduction is likely to be significantly less than the falls seen in financial wealth. 5
And it is important to remember that households may opt to mitigate the fall in their wealth not only through greater saving, but also by working more. Labour demand has, of course, weakened during the recession and with it the number of hours worked has fallen. But even if households are not able to increase their supply of labour now, they may plan to do so in the future, perhaps by delaying their eventual retirement. This may still help to moderate the extent to which spending falls in the near term.
3 Simple textbook models which tie consumption to permanent income would not predict this sort of behaviour. However, models which recognise imperfections in financial markets are often consistent with this type of behaviour, particularly when the effects of uncertainty are explicitly considered. For a discussion of consumption and saving theory see Berry, Waldron and Williams (2009). For a discussion of interactions between household debt, house prices and consumption, see Benito, Waldron, Young and Zampolli (2007).
4 For example, Bakhshi (2000) estimated that the value of future earnings (human wealth) represented around 80% of household wealth, whereas net financial wealth and gross housing assets together represented around 20%.
5 The sum of households' net financial and gross housing wealth has declined by over 10% since the start of the crisis. In contrast, external estimates of the loss of supply capacity as a result of the recession, to which the fall in human capital is likely to be closely related, are around the 4-5% range. See, for example, HMT (2009) and IMF (2009).
BIS Review 114/2009
None of this is to deny that households' balance sheets have been affected by recent falls in wealth. But the extent of that deterioration and hence the pressure on households to repair their balance sheets might be exaggerated by failing to take a more comprehensive view of their assets and liabilities.
Nevertheless, it is clear that, even putting balance sheets to one side, the financial crisis has affected households and families in ways that have caused them to spend less. Most directly, the recession has led to a pronounced weakening of the labour market, with employment falling and incomes squeezed.
The recession and the sharp rise in unemployment may also have led households to revise their beliefs about the stability of the economy and with it the predictability of their own incomes. The period from the early 1990s up to the financial crisis has been dubbed by economists as the Great Moderation. The UK economy enjoyed over 60 consecutive quarters of positive economic growth accompanied by low and stable inflation. It is possible that some households planned on the basis that this increased economic stability would persist indefinitely. If so, the events of the past two years would have come as a substantial shock.
Households' consumption has slowed sharply since the beginning of 2008. Indeed, consumption is estimated to have fallen for the past five consecutive quarters, the first time this has happened since records began in 1955. Relative to an average rate of increase, the level of consumer spending has fallen by around 7½% since the start of 2008, and the savings rate has increased sharply. There has already been a considerable adjustment in households' spending.
In the near-term, the prospects for consumption depend importantly on the performance of the labour market. Although there has been a substantial fall in employment over the past year or so, the size of this adjustment to date has, if anything, been less than we might have feared given the falls in output. This may partly reflect the greater degree of wage flexibility that has been apparent in this recession compared to that in either the 1980s or 1990s. 6 This greater flexibility has meant that more of the burden of firms' adjustment to the recession has been spread over the workforce as a whole, rather than on those losing their jobs. This may help to mitigate the slowing in consumption, both because households may find it easier to borrow to smooth their consumption if they still have a job and because it may limit the extent to which households increase their savings as a precaution against the possibility of future job losses.
Further out, the prospects for consumption will depend importantly on the extent to which the deterioration in households' financial position does indeed cause them to scale back their spending. As I have argued, there are good reasons for thinking this may happen. But it is also important to remember that there has already been a substantial adjustment in consumption; and that conventional measures of household balance sheets may exaggerate the scale of deterioration.
## Policy
Let me conclude with a few remarks on policy.
The current focus of monetary policy is on ensuring that there is sufficient stimulus in the economy to generate the strong and sustained recovery necessary to bring demand better into balance with supply and so ensure that inflation is on track to meet the inflation target. As I have said, the structural adjustments that need to occur - to the structure of the banking system; to the balance of global trade flows, and to the level of public and possibly
6 For more information see the box on page 29 of the August 2009 Inflation Report .
household debt - have the potential to delay and derail the recovery. But monetary policy can not - and should not - seek to prevent those adjustments. They are necessary for the long-run stability of our economy. Rather the job of monetary policy - and of public policy more generally - is to ensure that these adjustments occur in an orderly manner and within an economic environment which is consistent with hitting the 2% inflation target.
Because it is new and untested, there is understandably considerable nervousness about the Monetary Policy Committee's asset purchase programme - a.k.a. quantitative easing. How is it supposed to work? Will it work?
The essence of the asset purchase programme is to increase the supply of money and credit in the economy. An important point to stress is that the success of the programme does not depend on banks returning to a position in which they are able to lend normally. The policy works by supplying money directly into the economy, not through the banks. Much of the money is being injected by purchasing government bonds from institutional investors, such as insurance companies and pension funds. These purchases encourage investors to switch their portfolios away from gilts into corporate bonds and equities, so improving the supply of credit to businesses. Companies' access to credit is further enhanced by the Bank's purchases of commercial paper and corporate bonds, which improve the functioning of these markets.
But is it working? The objective of the asset purchases is to increase the growth of nominal spending. As with changes in interest rates, it will take some time before we see the impact of our purchases feeding through into higher aggregate spending and inflation. However, there are some encouraging signs that the medicine is working. The growth of underlying broad money has picked up in recent months, especially relative to the weak growth rates we might otherwise have seen. Interest rates have fallen. Corporate bond spreads have declined substantially. And the amount of finance raised by companies in bond and equity markets is at record levels. Not all of these developments can be attributed solely to our actions. But I have no doubt that asset purchases have played an important role.
At our policy meeting in August, the Monetary Policy Committee voted to maintain Bank Rate at 0.5% and to increase the scale of our asset purchase programme to £175 billion. That level of Bank Rate and scale of asset purchases were maintained at the MPC meeting earlier this month.
You may have notice that at our meeting in August there was a split vote amongst the Committee. A majority of the Committee - including myself - voted to increase the asset purchase programme to £175 billion, whereas a minority preferred a larger expansion to £200 billion. It is important not to make too much of this difference; all members were of the view that a further degree of monetary stimulus was needed to meet the inflation target.
My own judgment was that the uncertainty surrounding both the economic outlook and the effects of asset purchases made it difficult to have a strong view as to the relative merits of increasing the asset purchase programme to £175 billion or £200 billion. Indeed, the inflation projection that was published in the August Inflation Report suggested that either policy was consistent with hitting the inflation target, depending on the future path of interest rates. For me, the decision came down to an assessment of the risks of doing too much versus the risks of doing too little.
With regard to the potential risks of doing too little, my main concern was that nominal demand would not recover to a level consistent with hitting the inflation target. My main worry with doing too much stemmed from the uncertainty associated with using asset purchases as a policy instrument. We do not have much experience of conducting monetary policy via asset purchases and there is a risk that their effects will be transmitted through the economy in ways we do not predict and do not want. For example, the substantial injections of liquidity might result in unwarranted increases in some asset prices that could prove costly to rectify. I placed considerable weight on both risks and thought that they were best balanced by an expansion of the programme to £175 billion.
BIS Review 114/2009
I am hopeful that we have started along the road to recovery. The journey is likely to be long and not without difficulty. But the Monetary Policy Committee stands ready to do whatever it takes to ensure that the economy moves into better balance and inflation remains on track to hit the target.
## References
Bakhshi, H (2000), 'The sensitivity of aggregate consumption to human wealth', Bank of England Working Paper no. 108.
Benito, A., Waldron, M., Young, G. and Zampolli, F. (2007) 'The role of household debt and balance sheets in the monetary transmission mechanism' Bank of England Quarterly Bulletin Volume 47, no. 1, pp 70-78.
Berry, S., Waldron, M. and Williams, R. (2009) 'Household saving' Bank of England Quarterly Bulletin Volume 49, no. 3, pp 191-201.
HM Treasury (2009), 'Budget 2009, Building Britain's future'.
International Monetary Fund (2009), 'United Kingdom: 2009 Article IV Consultation'.
Sentence, A (2007), 'The changing pattern of savings: implications for growth and inflation', Royal Bank of Scotland / Scottish Economics Society lecture.