Striking a sensitive balance

Debt and what’s next

household consumption

Amit Kara examines what happens to economic growth if people stop borrowing or borrow only to fund necessary spending

READ NOW

Household consumption is the most important component of expenditure in UK national accounts. According to the Office for National Statistics (ONS), households spent some £1.36tn in 2019, which equates to an average of just under £600 per household per week.

The share of consumption in UK GDP is comparable with the US and is among the highest within the club of rich countries. It is no surprise then that UK economic growth is sensitive to the consumption behaviour of its households. It is also true that consumption spending is largely driven by household income, but history demonstrates that disruptions to consumer credit availability and demand have created large disturbances in the economy.

Historical data shows that the share of consumption in the UK economy has been broadly stable over time at around 60%, but its contribution to GDP growth can vary considerably year to year. Looking back at the period 1998-2019, household consumption contributed around two-thirds of annual GDP growth on average in these years.

This average, however, masks large swings from one period to another, so, for example, in the early period – 1998 to 2005 – consumption dominated the other components of spending (government, investment and trade), accounting for most of GDP growth. This was followed by the global financial crisis in 2008-09 when consumer spending was subdued and these other components drove GDP growth. What explains those swings?

Let’s start with some basic sector accounts. The households sector has two sources of funding – income and borrowing. The income category comprises wages and salaries, income from investments such as shares and property, and transfers, which is mainly government support through social security payments. A household can supplement its income by borrowing money from a variety of sources.

The combined resources of income and borrowing can either be used for consumption, to generate income or saved for future consumption. Most of consumption is funded by income. For example, the £1.36tn of consumption spending in 2019 is set against disposable income of £1.43tn in the same year. In other words, more than 90% of disposable income is consumed, which implies that the main driver for household consumption is disposable income.

We turn to borrowing next. Data from the Bank of England shows that UK households borrowed £61bn in 2019. Most of that, £48bn, was mortgage lending and the rest was unsecured lending, which comprises loans and credit card borrowing. A simplistic interpretation might go along the lines that if households had not borrowed the £61bn, GDP would have been lower by that amount in that year.

A home purchase involves additional spending, such as on furniture or white goods, which can be significant

That is not correct for several reasons. To begin with, most of that borrowing is for home purchase and a large proportion of that will fund the purchase of existing homes, which means that an existing asset is transferred to the buyer and the borrowed funds end up with the seller. There is no new good or service of that value that is produced. That said, a home purchase comes alongside additional spending on legal fees, furniture, white goods, etc, which can be significant.

The remaining £13bn was unsecured lending. Some of this would help service previous loans and most of the rest might fund consumption spending. This spending could include rent, utility bills, transport costs and other “non-fun’”items of expenditure. All in all, this £13bn accounts for around 1% of total consumption spending in 2019, which implies that the direct impact of zero unsecured borrowing is worth around 1% of consumer spending growth or around 0.6% of GDP.

Fall in non-mortgage borrowing

There have been two episodes in recent history when UK households largely stopped borrowing that was not in the shape of a mortgage. The first episode was around the global financial crisis and the second was the more dramatic shift in 2020, triggered by the Covid pandemic.

During the global financial crisis, lending more or less stagnated from 2008 to 2011 and household debt (as a percentage of income) started to fall. The economy slowed significantly, with average annual GDP growth at a negative 0.3% over this period.

The 2020 experience was more dramatic. Total lending to households decreased by £8bn in the second quarter of 2020, according to the Bank of England. Household consumption spending fell by more than £80bn, or 23.6%, over the same period, according to the Office for National Statistics.

Source Bank of England

This drop marks the largest quarterly fall in nominal household spending ever recorded. There were sharp falls in spending on restaurants and hotels, air transport and cars, and recreation and cultural services. GDP fell by a striking 20% in this quarter.

There are some distinct factors at play, but the one common driver for weak credit growth across both episodes is tighter lending conditions. Credit conditions for both secured and unsecured lending tightened in 2007 and 2008 and it wasn’t until 2012 that credit conditions started to ease significantly.

Credit conditions tightened once again in 2020 but policy intervention has helped reverse much of the tightening in the secured lending market. Lower borrowing costs and the temporary reduction in stamp duty helped the secured lending market to bounce back impressively in September and October last year to levels well above the 2019 average.

Does UK growth rely on consumer borrowing?

Most of consumption is funded by household disposable income, yet the economy is heavily dependent on a well-functioning credit market. The two episodes discussed above highlight the short-term impact of borrowing on consumption and GDP, but this is not the complete picture. Borrowing allows households to smooth consumption spending over a period of time. A good example of smoothing is mortgage lending. Households use a mortgage to secure an asset during their working life that they can then use during retirement when income is typically much lower.

Note: Positive balances indicate that lenders, on balance, reported/expected demand/credit availability/defaults to be higher than over the previous/current three-month period, or that the terms and conditions on which credit was provided became cheaper or looser respectively.

An abrupt disruption to the mortgage market will have a direct impact on housing transactions, house prices, consumption and housebuilding. Although the housing market per se is relatively small, the two episodes of credit tightening came alongside a sharp slowdown in economic activity. Had credit conditions not improved, housing transactions and investment would have stalled and house prices would have become permanently lower, with a knock-on effect on household consumption and government revenue and spending.

The recession was deepest in those countries that had rapid credit expansion and where indebtedness was high

Sustainable economic growth requires responsible borrowing and lending. With the benefit of hindsight, we know that credit growth was unsustainably quick in the three to four years leading up to the global financial crisis in 2008. That trend reversed rapidly during the crisis when households looked to restore their balance sheet by spending less and lowering indebtedness. We also know from research, by Bunn and Rostom at the Bank of England, that the recession was deeper in countries that had experienced rapid credit expansion and where indebtedness was high.

How much should households borrow?

Most people will agree that households will have a minimum amount of spending that must include adequate food, a decent home, clothing, recreation, education, etc. But what constitutes necessary spending even within that is a tricky question.

A spiritually minded person will argue that “less is more”, but not everyone has a spiritual bent. Does what is essential include the purchase of a home or a second home, a car when public transport is available, a foreign holiday, etc?

An alternative question at the macroeconomic level might be, what is the optimum level of borrowing? UK household debt has doubled (as a share of GDP) over the past 40 years. Is that sustainable? One of the key reasons for rising indebtedness is the secular decline in inflation and the real interest rate over this period. Debt has become more affordable because the cost of servicing that debt has fallen thanks to lower real interest rates.

The cost of borrowing fell once again in 2020 when the Bank of England lowered Bank rate to just 0.1%. The current debt levels at the aggregate level appear sustainable at the present borrowing costs, but that could change if, for example, a supply shock drives inflation and inflation expectations rise sharply, forcing the Bank of England to raise borrowing costs. In other words, borrowing and spending that are affordable today could easily turn unaffordable under a different inflation and policy rate regime. Households, lenders and regulators must remain alert to that risk.

Amit Kara

Amit Kara is an independent consultant with a special interest in global economics and climate change. He worked at the National Institute of Economic and Social Research most recently, where he helped develop the climate change version of the National Institute Global Econometric Model. He has also worked for the Bank of England, at UBS as chief UK economist, and at HSBC