GDP growth is expected to have recovered in 2018 Q2, having slowed temporarily in Q1. Real income growth is recovering following the effects of sterling’s depreciation, which should support modest consumption growth. Business investment and net trade should also continue to support GDP growth, though remain sensitive to the global outlook and the effects of Brexit.
2.1 Near-term outlook
Quarterly GDP growth is estimated to have slowed to 0.2% in 2018 Q1 (Chart 2.1). That was revised up from 0.1% in the preliminary estimate and, as set out in the May Report, it is expected to be revised up further to 0.3% in the mature estimate.
In May, the MPC judged that growth in Q1 was probably depressed by around 0.1 percentage points by disruption from adverse weather. Developments since then have been broadly consistent with that judgement. For example, according to Bank calculations based on responses to the ONS Labour Force Survey, total hours worked were 0.15% lower in Q1 due to the adverse weather.
GDP growth is expected to have recovered to 0.4% in Q2 (Chart 2.1), as anticipated in May. That is slightly faster than the estimated growth rate of potential supply — the pace at which output can grow consistent with balanced inflationary pressures. Newly introduced ONS estimates of monthly GDP growth (see Box 2) suggest that growth in the three months to May was 0.2%. That growth rate continued to be depressed by the impact of weak activity in March however, probably due to the adverse weather. By contrast, monthly growth in April and May averaged ¼%.
The recovery in GDP growth in Q2 is expected to have been driven by a pickup in consumption growth, to 0.5% (Section 2.2). A number of indicators of household spending, including consumer credit growth and property transactions (Section 2.3), which were weak in Q1, have bounced back since then, suggesting much of the earlier weakness was erratic. In addition, retail sales grew by 2.1% in Q2 (Chart 2.2). Although in the past year the number of retail store closures have increased and retail footfall has fallen, contacts of the Bank’s Agents suggest that mainly reflects shifts in consumer demand to online stores and from goods to services. And although growth in household money has slowed, that appears to reflect an unwind of past shifts in demand for different assets (see Box 3).
In contrast to consumption, net trade is expected to have subtracted from growth in Q2, in part due to a fall in goods exports (Section 2.4). Consistent with that, manufacturing output has weakened since the start of 2018 (Chart 2.3), although part of that recent fall could reflect a lagged impact from the weather-related disruption in Q1. Companies built up inventories in the first quarter (Table 2.A), but surveys suggest that inventories fell back in Q2 and so companies should need to raise activity to meet further demand growth. Net trade is projected to contribute positively to growth in subsequent quarters.
GDP growth is projected to remain at 0.4% in Q3 (Chart 2.1). Most survey indicators of output remain consistent with steady growth over the rest of 2018.
GDP growth is expected to have picked up in Q2 following temporary weakness in Q1
Output growth and Bank staff’s near-term projection(a)
- Sources: ONS and Bank calculations.
(a) Chained-volume measure. GDP is at market prices. The diamonds show Bank staff’s projection for the first estimate of GDP growth in 2018 Q2 and Q3. The bands on either side of the diamonds show uncertainty around those projections based on the out-of-sample performance of Bank staff’s best performing model since 2004, representing ±1 root mean squared error (RMSE). The RMSE of 0.1 percentage points around the 2018 Q2 projection excludes three quarters affected by known erratic factors: the 2010 snow and the 2012 Olympics and Diamond Jubilee. Including those erratic factors, the RMSE for 2018 Q2 rises to 0.2 percentage points. For 2018 Q3, the RMSE of 0.3 percentage points is based on the full evaluation window.
Retail sales growth rose sharply in Q2
Retail sales volumes and survey indicators of retail sales
- Sources: Bank of England, British Retail Consortium (BRC), CBI, ONS, Visa and Bank calculations.
(a) Chained-volume measure.
(b) Swathe includes: BRC percentage change in total sales, not seasonally adjusted; balance of respondents to the CBI distributive trade survey question ‘How do your sales and orders this month compare with a year earlier?’; percentage change in Visa total consumer spending on a year ago, deflated by CPI inflation; Agents measure of companies’ reported annual growth in the value of retail sales over the past three months, monthly measure until August 2016 and six weekly thereafter. All series have been scaled to match the mean and variance of ONS retail sales volumes growth since 2000 except the BRC and CBI series, which are since 1995, and the Visa series, which is since 2006.
Manufacturing output has weakened since the start of 2018
Manufacturing output and survey indicators of manufacturing activity
- Sources: Bank of England, BCC, CBI, EEF, IHS Markit, ONS and Bank calculations.
(a) Swathe includes: Markit/CIPS net percentage balance of manufacturing companies reporting that production/output increased this month compared with the previous month; CBI average of the net percentage balances of manufacturing companies expecting the volume of output to rise; EEF average of the net percentage balances of manufacturing companies reporting that total output was expected to increase over the next three months; BCC net balances of manufacturing companies that reported that turnover will improve over the next 12 months; Agents measure of companies’ reported annual growth in the value of manufacturing output over the past three months. Scaled to match the mean and variance of four-quarter manufacturing output growth since 2000.
(b) Chained-volume measure at basic prices. Three-month moving average.
Expenditure components of demand(a)
- (a) Chained-volume measures unless otherwise stated.
(b) Includes non-profit institutions serving households (NPISH).
(c) Investment data take account of the transfer of nuclear reactors from the public corporation sector to central government in 2005 Q2.
(d) Excludes the alignment adjustment.
(e) Percentage point contributions to quarterly growth of real GDP.
(f) Includes acquisitions less disposals of valuables.
(g) Excluding the impact of missing trader intra-community (MTIC) fraud.
2.2 Consumer spending
Consumer spending is determined largely by household incomes. Household real income growth has been weak since 2016 (Chart 2.4) due to rises in import prices following the depreciation of sterling associated with the EU referendum and subdued nominal pay growth over that period. As the effects of the fall in sterling on inflation fade and nominal pay growth continues to pick up, real income growth is expected to be higher than in recent years, supporting consumption growth (Section 5).
Over the recent past, consumption growth has slowed by less than real income growth and, as a result, the saving ratio has fallen since the end of 2015 (Chart 2.4). The extent to which households choose to save or borrow is likely to depend in part on the state of their balance sheets, which are relatively strong compared to the past (see Box 4).
Another factor that can affect how much households spend or save is their confidence in future income growth. The headline GfK measure of consumer confidence has been broadly stable since mid-2016 (Chart 2.5). Within that, the balance for households’ expectations of their personal situation, which tends to be well correlated with spending growth, has risen in recent months. Consumer confidence is likely to have been supported by the strength of the labour market (Section 3) with the latest IHS Markit Household Finance Index suggesting that job security was around its highest level since the series began in 2009.
Household spending will also be influenced by interest rates. First, changes in how much income households receive from their deposits or are required to pay on their debts will affect how much they have available to spend. Net savers are typically less inclined to spend out of a rise in income than net borrowers and so rises in interest rates tend to depress spending overall. Second, interest rates can influence how much of their incomes households choose to save, or how much they borrow to supplement their income, with higher interest rates increasing the attractiveness of saving and reducing that of spending, all else equal.
In the mortgage market, lower bank funding costs and intensifying competition have led to a reduction in mortgage rates in recent years. Some households will have benefited from those low rates either by taking out new mortgages or by remortgaging, which all else equal will have boosted household spending. Bank funding costs have risen more recently (Section 1), however, and are starting to be passed through to higher mortgage rates.
In consumer credit, there is evidence of a modest tightening in conditions over the past year. The maximum interest-free period on credit card balance transfers and on purchases, on average across lenders, have both continued to fall, while the average quoted rate on credit cards has risen. Consumer credit conditions remain supportive, however, and results from the latest Bank/NMG survey show that the share of respondents concerned about their access to credit remains low.
Taking these influences together, consumption is expected to grow modestly in coming quarters, at around ¼% on average (Table 2.B), broadly in line with real incomes and supported by accommodative financial conditions. That outlook is corroborated by results of a survey of companies on consumer demand conducted by the Bank’s Agents in May, where respondents expected stronger consumer income growth and confidence to drive an acceleration in sales volumes over the next 12 months.
Consumption growth has slowed by less than real income growth so the saving ratio has fallen
Consumption, real post-tax income and household saving(a)
- Sources: ONS and Bank calculations.
(a) Shaded area denotes the period following the EU referendum. All series include NPISH.
(b) Saving as a percentage of household post-tax income.
(c) Nominal post-tax income divided by the consumption deflator.
(d) Chained-volume measure.
Consumer confidence has been broadly stable since mid-2016
Indicators of consumer confidence
- Sources: GfK (research carried out on behalf of the European Commission) and Bank calculations.
(a) Average of the net balances of respondents reporting that: their financial situation has got better over the past 12 months; their financial situation is expected to get better over the next 12 months; the general economic situation has got better over the past 12 months; the general economic situation is expected to get better over the next 12 months; and now is the right time to make major purchases, such as furniture or electrical goods.
Monitoring the MPC’s key judgements
Developments in the housing market can provide a signal about household spending more generally. Over the past few years, activity in the housing market has been broadly stable, but subdued. Mortgage approvals have been broadly unchanged since mid-2016 despite low interest rates (Chart 2.6). Annual house price inflation has slowed since the start of 2016, by around 5 percentage points, to 2.2% in 2018 Q2 according to the average of lenders’ indices (Chart 2.7), slower than expected at the time of the May Report. Consistent with that, official UK data to May suggest house price inflation has declined to a similar degree. Rent inflation has also slowed and was around 1% in the year to 2018 Q2 (Chart 2.8).
That weakness in the housing market appears to be concentrated in London. In 2018 Q1, mortgage completions for housemovers and first-time buyers in London were around 12% lower than in 2016, and house price (Chart 2.7) and rent inflation (Chart 2.8) have both fallen sharply and are now negative.
Although developments in London have tended to lead other areas in the past, if the reasons for the current weakness in London are fairly idiosyncratic, they may indicate little about prospects for the UK housing market as a whole. London house price inflation was particularly strong from 2014–16 (Chart 2.7), reducing affordability. Given its relatively high level of house prices, London was likely to have been disproportionately affected by regulatory and tax changes since 2014. The recent slowing in house prices has brought London more into line with other areas. The slowing in the buy-to-let market, as mentioned in previous Reports, may have affected London more than other areas, as London accounts for a substantial proportion of UK buy-to-let activity.
Some of the weakness in the London market may also reflect a fall in net EU migration (Section 3), alongside wider Brexit uncertainty. The number of EU nationals in London appears to have fallen slightly since the EU referendum, although it has continued to grow in other regions. As EU nationals make up around 12% of households in London and 20% of the private rental sector, that fall will affect demand for housing services and therefore house prices and rents. Alongside that, more respondents to the RICS housing survey in London than in other areas reported increased uncertainty about the UK’s future relationship with the EU and an associated negative impact on house prices.
In the near term, modest real income growth and accommodative credit conditions should support housing market activity. Mortgage approvals are projected to remain stable and UK house price inflation is expected to pick back up to a little over 3% by mid-2019 (Table 2.B).
Developments in the housing market will also contribute to GDP directly through housing investment. Housing investment fell by 0.5% in 2018 Q1, but that may in part reflect adverse weather, which hampered construction activity. Construction activity is expected to have recovered in Q2, although contacts of the Bank’s Agents report that capacity constraints were limiting the extent to which any output lost in Q1 could be subsequently made up. New housing orders, which are an indicator of private housing starts and therefore investment, rose by 19% in 2018 Q1 to their highest level since the crisis. This will support housing investment over the rest of the year.
Mortgage approvals for house purchase have been stable but subdued
House price inflation has slowed particularly sharply in London
- Sources: IHS Markit, Nationwide and Bank calculations.
(a) Averages of the quarterly Halifax/Markit and Nationwide house price indices.
Slowing rent inflation has been largely concentrated in London
Private housing rent(a)
- (a) Not seasonally adjusted. Data for Northern Ireland are not available.
2.4 Net trade and the current account
Net trade contributed positively to GDP growth in 2017 and continued to do so in 2018 Q1 (Table 2.A). In contrast, net trade is expected to have subtracted 0.6 percentage points from GDP growth in Q2. However, the recent fall in part reflects volatility in net exports of non-monetary gold, which do not affect aggregate GDP as they are offset by changes in the contribution to private sector investment in valuables. Survey indicators of export growth remain robust (Chart 2.9) and demand for exports will continue to benefit from relatively robust global growth (Section 1). The outlook for net trade will depend on how the supply chains and capacity of companies, both here and abroad, evolve in response to Brexit and associated movements in sterling. But net trade is expected to continue to make a positive contribution to GDP growth in coming quarters.
The current account deficit — which reflects the balance of nominal trade flows and other payments between the UK and the rest of the world — narrowed to 3.4% of GDP in 2018 Q1 (Chart 2.10). That reflected a narrowing in the deficits on both the trade balance and the primary income balance — the net value of investment income received by UK residents. Over 2008–17, the trade deficit is estimated to be around 0.4% of GDP narrower on average than previously estimated. That reflects revisions to data in Pink Book 2018 following a methodological change that increased the estimated level of services exports.
UK export growth appears to have slowed in Q2 although survey indicators remain robust
UK exports and survey indicators of export growth
- Sources: Bank of England, BCC, CBI, EEF, IHS Markit, ONS and Bank calculations.
(a) Swathe includes: BCC net percentage balance of companies reporting that export orders and deliveries increased on the quarter (data are not seasonally adjusted); CBI average of the net percentage balances of manufacturing companies reporting that export orders and deliveries increased on the quarter, and that their present export order books are above normal volumes (the latter series is a quarterly average of monthly data); Markit/CIPS net percentage balance of manufacturing companies reporting that export orders increased this month compared with the previous month (quarterly average of monthly data); Agents measure of manufacturing companies’ reported annual growth in production for sales to overseas customers over the past three months (last available observation for each quarter); EEF average of the net percentage balances of manufacturing companies reporting that export orders increased over the past three months and were expected to increase over the next three months. Indicators are scaled to match the mean and variance of four-quarter export growth since 2000.
(b) Chained-volume measure, excluding the impact of MTIC fraud. The diamond shows Bank staff’s projection for 2018 Q2.
The current account deficit narrowed in Q1
UK current account
2.5 Business investment
Business investment fell by 0.4% in 2018 Q1 (Table 2.A). That fall was driven by investment in construction-related assets, which may have reflected the effect of adverse weather. Even looking through that volatility, however, business investment growth has continued to be weaker in recent years than in previous recoveries and lower than would be expected given accommodative financial conditions and relatively robust global growth.
Some of this recent weakness may reflect the effects of uncertainty around Brexit. As set out in the May Report, results from the Bank’s Decision Maker Panel Survey and Deloitte CFO Surveys at that time suggested that Brexit was becoming less of a drag on business investment growth. But respondents to the 2018 Q2 Deloitte CFO Survey again ranked Brexit as the top risk facing their businesses, and three quarters of respondents expected Brexit to lead to a deterioration in the business environment in the long term, the highest proportion since the referendum.
Weak demand for investment appears to have been reflected in slowing growth of bank lending to companies since mid-2016 (Chart 2.11). There are also signs that financial conditions have tightened slightly since May which might have weighed on lending, although conditions remain accommodative overall (Section 1). Growth in lending to small and medium-sized enterprises has been slower than for larger companies. Results from the Q2 Credit Conditions Survey, however, suggested that demand for bank lending from small businesses rose significantly in that quarter.
Larger corporates can also access other sources of finance, such as the corporate bond market. Net corporate bond issuance was strong in Q2 (Chart 2.12). Additionally, the volume of leveraged loans has risen rapidly since early 2017. But over that period, almost all of those loans have been for either balance sheet restructuring or mergers and acquisitions and so are unlikely to have provided much direct support to business investment growth.
Business investment growth is expected to have picked back up in Q2 and to remain a little above its past average rate in coming quarters (Table 2.B), but subdued compared with previous recoveries. That is consistent with survey measures of investment intentions. Investment is likely to remain sensitive to developments in Brexit and the prospects for global growth.
Growth in bank lending to companies has slowed
Lending to UK non-financial businesses(a)
- (a) Rate of growth in the stock of lending by UK monetary financial institutions (MFIs) in all currencies, expressed in sterling. Not seasonally adjusted.
(b) Large businesses are those with annual debit account turnover on the main business account of over £25 million.
(c) Small and medium-sized enterprises are businesses with annual debit account turnover on the main business account of equal to or less than £25 million.
Net corporate bond issuance picked up in Q2
Net external finance raised by UK private non-financial corporations(a)
- (a) Includes sterling and foreign currency funds from UK MFIs and capital markets.
(b) As component series are not all seasonally adjusted, their sum may not equal the total.
(c) Includes stand-alone and programme bonds.
(d) Not seasonally adjusted.
The MPC’s projections are conditioned on the Government’s tax and spending plans detailed in the March 2018 Spring Statement. Under those plans, public sector net borrowing is projected to fall to 1.3% of GDP by 2020/21, from 2.3% of GDP in 2017/18.
Box 2: Implications of changes to the ONS GDP publication timetable
Following a consultation, the ONS announced improvements to the coverage of GDP estimates and changes to the timing of releases. Those changes are beginning to be introduced, altering the presentation of data for 2018 Q2. This box discusses how the new GDP release schedule affects the MPC’s monitoring of the current economic conjuncture. Overall, the changes have no material implications for the extent of the information available to the MPC at its policy meetings.
The new ONS timetable
To improve the accuracy and reliability of the first estimate of GDP, the ONS is delaying the release by two weeks (Figure A), allowing it to incorporate a greater amount of data. While that takes publication beyond the timing of Inflation Reports, another welcome development is that monthly estimates of output in the service sector — which constitutes around 80% of the economy — will be published earlier, alongside those for production and construction. That means output data for the first two months of the latest quarter will still be available ahead of each Report, and summarised in a new monthly GDP time series. While estimates of the third month will not be available until later, the data content of those in previous preliminary estimates was significantly lower than for the other two months.
Since official data are published with a lag, the MPC produces estimates of the current rate of GDP growth, or ‘nowcasts’. While such nowcasts are subject to judgement, they are heavily informed by a range of models.1 One of these models, using a mixed-data sampling (MIDAS) approach, is particularly suited to nowcasting. This model takes into account a range of survey and official data, with the weights attached to the various indicators changing as more data become available. Early estimates place a high weight on the latest surveys, which tend to be more timely, with greater emphasis placed on official data as they become available.
Bank staff nowcasts should provide a good signal for early estimates of quarterly GDP growth. Testing their past performance, by feeding in two months of real-time output data, suggests that GDP nowcasts using Bank staff’s latest MIDAS model would have been within 0.1 percentage points of the ONS’s preliminary estimate on around 80% of occasions since 2004 (Chart A).
Moreover, GDP data are revised over time as a wider range of information becomes available and methodological improvements allow the ONS to measure activity more accurately. Uncertainty around the profile of GDP growth will, therefore, continue to exist beyond the release of early estimates.
1. For further details, see Anesti, N, Hayes, S, Moreira, A and Tasker, J (2017), ‘Peering into the present: the Bank’s approach to GDP nowcasting’, Bank of England Quarterly Bulletin, 2017 Q2. Bank staff have since introduced a new MIDAS model, which is better suited to the ONS’s new publication timetable.
Some releases have been brought forward, with an official estimate of quarterly GDP growth published later
GDP release schedules
Nowcasts based on two months of official data are close to early estimates of quarterly growth
Nowcasts and preliminary estimates of GDP
- Sources: BCC, CBI, IHS Markit, ONS and Bank calculations.
(a) Estimates from Bank staff’s latest MIDAS model, using survey data and real-time ONS data for the first two months of output for the quarter.
Box 3: Developments in broad money
Money is the key medium of exchange with which to make payments for goods and services.1 As such, money growth may provide a signal for recent and future trends in activity and inflation.2 There are a number of factors that affect the amount of money in circulation, however, and understanding why money growth has evolved as it has is important in assessing what signals to take from it. This box considers recent developments in broad money — the amount of money held in bank deposits and as cash in circulation.
Twelve-month growth in broad money slowed to 3½% in 2018 Q2, having been above 7% in 2016 H2 (Chart A). That has brought the rate closer to nominal spending growth, which has also slowed, albeit to a lesser degree. A slowdown in credit growth since 2016 is likely to have been a contributing factor to the slowing in money growth as it is the main source of money creation.3 Indeed, the two have slowed by a similar extent.
Looking beneath the aggregate data, developments in money holdings across different sectors of the economy may also help to highlight trends. Those developments can, however, also reflect other factors that have limited implications for spending prospects, which appears to have been the case recently. And since the sectors are interconnected, developments in one sector can also spill over to others as money circulates around the economy.
The slowing in aggregate broad money growth since 2016 largely reflects slower growth in households’ deposits and cash holdings (Chart B). Household spending growth has slowed at the same time, albeit by slightly less. Although household spending and money growth are correlated, other household indicators (Section 2.2) have statistically been better predictors of spending, and money balances have provided little incremental information over and above those.
As discussed in past Reports, the pickup in household money growth during 2016 occurred alongside a reduction in households’ investment fund holdings and so appeared to reflect a desire among households to hold more liquid assets in the face of heightened uncertainty around the referendum. The subsequent slowing in household money growth appears in part to reflect some of that precautionary demand subsiding, as investment fund holdings have risen. To the extent that it reflects a shift in demand for different assets, its effect on money growth is likely to be temporary and to be less informative about spending prospects.
Private non-financial companies’ (PNFCs’) deposits have in the past appeared to be a leading indicator of business investment growth.4 Growth in PNFCs’ money holdings has been robust in recent years, having picked up since 2011 (Chart B). But that in part appears to have reflected an increase in companies’ desired money holdings — continuing a trend that emerged prior to the crisis — perhaps for precautionary purposes. As such, nominal business investment growth continues to be modest despite that growth in companies’ money holdings. Nevertheless, those higher balances may provide some support to future business investment growth.
Money holdings of non-intermediate financial companies — covering pension funds and other asset managers — may influence, as well as be influenced by, asset prices.5 Many of the MPC’s asset purchases, announced in August 2016, will have been bought from asset managers and hence boosted their money holdings at the time. To the extent that their money holdings were above desired levels as a result, they may have bought other assets to rebalance their portfolios, boosting the prices of those assets.6 For example, growth in equity prices appear to have been correlated with growth in money holdings (Chart B). The impact on equity prices of those and earlier asset purchases by the Bank are difficult to detect, however, given the many other factors that affect them.7 For example, the decline in the exchange rate during 2016 boosted the sterling value of profits earned in UK-listed companies’ overseas operations, and therefore sterling equity prices (Section 1).
Overall, money growth can be affected by a number of factors that complicate its relationship with activity and prices. Underlying trends in money holdings do tend to broadly coincide with nominal spending, however, and as a timely indicator of economic developments, the MPC will continue to monitor them closely alongside other indicators of spending.
1. For further discussion of the role of money in the economy see Carney, M (2018), ‘The future of money’.
2. See for example McLeay, M and Thomas, R (2016), ‘Broad money growth in the long expansion, 1992–2007: what can it tell us about the role of money?’, in Chadha, J, Chrystal, A, Pearlman, J, Smith, P and Wright, S (eds), The UK economy in the long expansion and its aftermath, Cambridge University Press.
3. For further detail on how money is created see McLeay, M, Radia, A and Thomas, R, (2014), ‘Money creation in the modern economy’, Bank of England Quarterly Bulletin, 2014 Q1.
4. See for example Brigden, A and Mizen, P (2004), ‘Money, credit and investment in the UK industrial and commercial companies sector’, The Manchester School, Vol. 72, No. 1, pages 72–79.
5. See for example Congdon, T (2005), ‘Money and asset prices in boom and bust’, Institute of Economic Affairs, Hobart Paper No. 152.
6. For further detail on the link between asset purchases and the broad money stock, see Butt, N, Domit, S, McLeay, M and Thomas, R (2012), ‘What can the money data tell us about the impact of QE?’, Bank of England Quarterly Bulletin, 2012 Q4.
7. For further discussion see Broadbent, B (2018), ‘The history and future of QE’.
Broad money growth has slowed since 2016, as have credit and spending growth
Broad money, nominal GDP and credit
- (a) M4 and M4 lending excluding the deposits of, and borrowing by, intermediate other financial corporations.
(b) At current market prices. Data are to 2018 Q1.
Household money growth has fallen below spending growth, while corporate money growth continues to exceed investment growth
Sectoral broad money, spending and equity prices
- (a) Monetary financial institutions’ sterling M4 liabilities to the respective sectors.
(b) At current market prices. Data are quarterly to 2018 Q1.
(c) End-quarter observations.
Box 4: Household balance sheets
Household real income growth has been weak due to the rise in import prices following sterling’s depreciation around the time of the EU referendum and subdued nominal pay growth. Real incomes have been broadly flat since the start of 2016, compared to average quarterly growth of ½% during 2012–15. Consumption growth has not slowed to the same extent as real income growth and, as a result, the saving ratio has fallen to historically low levels (Chart 2.4).
A key influence on household spending will be the pace at which households might seek to build savings in coming years. That will depend, in part, on the strength of household balance sheets. Households may be willing to save less if they have already built up a buffer stock of precautionary saving, or if the value of their net wealth rises. But they might save more if they have a low stock of saving and are worried about their ability to borrow to support consumption, in case of future loss of income. This box explores developments in household balance sheets and discusses the potential implications for savings.
Net wealth and household saving
The stock of households’ financial and housing assets is considerably larger than their stock of debt. That means, in aggregate, the household sector holds significant net wealth. Developments in household balance sheet positions will reflect patterns of asset and liability accumulation as well as changes in asset values.
In the decade prior to the financial crisis, net wealth rose steadily (Chart A) and the saving ratio was broadly stable. But during the crisis, wealth fell, driven mainly by a fall in house prices, and the saving ratio rose. Some of that rise in the saving ratio may have reflected households increasing saving to offset the fall in wealth. Some of it is also likely to have reflected the sharp tightening in credit conditions and higher uncertainty associated with rises in unemployment over that period.
Since the crisis, net financial wealth has risen by around 60% (Chart A), credit conditions have loosened, unemployment has decreased and the saving ratio has fallen back. In particular, since the end of 2015, net financial wealth has risen by 12%, in part as the fall in sterling associated with the EU referendum has boosted the value of equities of companies with foreign earnings (Section 1), and the saving ratio has fallen further, to historically low levels.
The extent to which rises in wealth boost consumption and lower the saving ratio is uncertain and can vary over time. Bank staff estimate that, on average over the past, a 10% rise in the real value of households’ financial assets has boosted consumption by around 0.5%. On that basis, the rise in real net financial wealth since the end of 2015 could have been associated with around a ½ percentage point fall in the saving ratio.
In addition to the rise in financial wealth, housing wealth has risen by around 12% since the end of 2015 and by around 40% since the crisis (Chart A). Rises in the value of housing wealth will increase the value of housing equity that households can use as collateral against which to borrow. As set out in the box on pages 18–19 of the November 2016 Report, Bank staff estimated that for a 10% rise in housing wealth, this collateral channel was associated with a boost to the level of consumption averaging around 0.5%.
The distribution of assets and debt across households
The distribution of assets and liabilities across households can also matter for consumption. In particular, households with higher levels of debt tend to adjust their spending more significantly in response to shocks to their income.1
Results from the latest Bank/NMG survey suggest that the proportions of households with high mortgage debt to income ratios or high debt-servicing ratios (DSRs) have risen slightly since 2016, but remain significantly lower than during the crisis (Chart B). The proportion of households with high DSRs remains below its pre-crisis average. In addition, the savings of households with high DSRs appear to have risen since 2014 (Chart C), which suggests that those households are now likely to be more resilient to shocks.
Results from the latest Bank/NMG survey also show that more households with high levels of net financial wealth have maintained or increased their spending over the past 12 months than less wealthy households (Chart D). While those households may have seen faster rises in their incomes, it could suggest that having stronger balance sheets has also supported their spending.
In aggregate, household balance sheets look relatively strong compared to the past as the stock of both financial and housing assets has risen considerably faster than the stock of debt. That may be supporting the current low saving rate, allowing households to maintain consumption growth even as real income growth has slowed.
Alongside that, low unemployment and accommodative financial conditions should limit households’ perceived need to build up further precautionary savings in aggregate. As a result, consumption growth is expected to remain in line with real income growth over the next few years such that the saving ratio remains broadly flat (Section 5).
1. For more detail see Bunn, P and Rostom, M (2014) ‘Household debt and spending’, Bank of England Quarterly Bulletin, 2014 Q3.
Net wealth has risen since the financial crisis
Net household wealth
- Sources: IHS Markit, Nationwide, ONS and Bank calculations.
(a) Financial wealth less unsecured debt. Not seasonally adjusted.
(b) Housing wealth less secured debt. Housing wealth from 2017 onwards is an estimate based on growth in UK house prices according to the average of the quarterly Halifax/Markit and Nationwide house price indices. Not seasonally adjusted.
(c) Sum of net financial and net housing wealth.
The proportion of households with high mortgage debt relative to income is lower than during the crisis
Proportions of households with high mortgage debt relative to gross income and a high mortgage DSR(a)
- Sources: British Household Panel Survey (BHPS), Living Costs and Food (LCF) Survey, NMG Consulting survey, Understanding Society Survey (USS) and Bank calculations.
(a) Calculated using LCF Survey (for debt to income ratios) and BHPS and USS (for debt-servicing ratios) from 1997 to 2014, shown in the dashed lines, and NMG Consulting survey from 2012 onwards, shown in the solid lines.
(b) Debt to income calculated as gross mortgage debt as a percentage of a four-quarter moving sum of disposable income.
(c) Mortgage DSR calculated as total mortgage payments as a percentage of pre-tax income.
Households with high mortgage DSRs have more savings than in previous years
Median savings of households with high mortgage DSRs(a)
- Sources: NMG Consulting survey and Bank calculations.
(a) Mortgage DSR calculated as total mortgage payments as a percentage of pre-tax income. High DSRs defined as 40% or greater. Savings calculated using answers to the following questions in the NMG survey: ‘How much do you (and all other members of your household) currently have in total, saved up in savings accounts? Please include bank/building society accounts or bonds, cash ISAs, and NS&I account/bonds.’; and ‘Excluding money in savings accounts, how much do you (and all other members of your household) currently have in total, saved up in other investments such as stocks, shares and unit trusts?’
More households with high levels of wealth have maintained or increased spending than less wealthy households
Changes in household spending over the past 12 months by wealth decile(a)
- Sources: NMG Consulting survey and Bank calculations.
(a) Wealth decile defined by total gross financial wealth. Higher deciles indicate higher levels of wealth. Changes in household spending calculated using responses to the question in the NMG survey: ‘How has your household changed its spending compared with a year ago? Please include all your spending on goods and services, but exclude money put into savings or used to repay mortgages, overdrafts, credit cards and other loans.’