Global Monthly - Will cash buffers limit the coming downturn?

Global View: Buffers may delay the economic pain necessary to lower inflation, but not avoid it

Advanced economies seem to be caught in a tug-of-war recently between a resumption of the post-pandemic rebound on the one hand, and on the other, tighter financial conditions resulting from aggressive rate hikes and the recent banking turmoil. April flash PMIs for both the eurozone and the US suggest continued economic resilience – almost entirely concentrated in services – while other surveys, including the US NFIB and the Fed's Beige Book point to a building headwind from tighter bank lending standards. To add to the mix of conflicting drivers, China's reopening rebound is also turning out to be stronger than expected. We judge that the recent resilience in the US and eurozone will prove temporary, and that tighter financial conditions will ultimately prevail, pushing advanced economies into recession.

In this month's Global View, we the explore the link between credit and GDP growth historically, and assess whether the strength in household and corporate balance sheets may partly explain why tighter financial conditions are taking longer to affect activity. The rebound in momentum suggested by the PMIs is not a positive for the medium-term outlook, in our view. Rather, it raises the risk that central banks may have to go further in hiking rates to ensure inflation falls fully back to target.

Taking stock: Financial conditions – and the risk of a credit crunch – have eased

Since our March publication, worries over the banking sector have significantly receded. Flows to money market funds in the US have gone into reverse, the banking sector in aggregate has seen a recovery in deposits, and weekly lending data has stabilised. Financial market sentiment has been supported by a slew of Q1 bank earnings releases which showed that small and medium banks have managed to navigate the turbulence so far. Indeed, Bloomberg’s US financial conditions index has fully retraced the tightening that took place in March and is now well into accommodative territory. Reflecting this, daily issuance of short-term corporate debt (commercial paper) – which briefly collapsed in March – is now back to normal.

Set against these positive developments, more qualitative or anecdotal evidence does point to some tightening in bank lending conditions. The small business NFIB survey suggested a significant decline in loan availability, with this sub-index falling to an 11 year low in March. The Fed’s April Beige Book last week also pointed to increased difficulty in acquiring credit. A key focus for policymakers will be the quarterly updates to the Fed’s Senior Loan Officer’s Survey (SLOOS) and the ECB’s Bank Lending Survey (BLS), due out in the coming weeks. These are likely to show some further tightening in lending standards, though it remains to be seen whether such a tightening will be sustained.

As we described in our March Monthly, a major tightening in lending standards poses significant downside risks to the growth outlook, as it would do part of the central bank’s work in terms of bearing down on credit growth and in turn economic activity. Indeed, we find that lending standards surveys are a strong leading indicator for economic activity, and this is something we explore in Box 1.

We still expect recessions in the US and eurozone – buffers can only delay the timing

Given the tightening in financial conditions – primarily driven by the surge in interest rates – we have long expected that the US would experience a mild recession in 2023, and the consensus and more recently the Fed’s board staff have also come around to this view. In the eurozone, too, although the near-term outlook has improved on the back of the receding energy crisis and the rebound in China (see Box 3), we still expect a moderate recession this year. However, a key argument against these mild/moderate expected downturns becoming more severe is the strength in household and corporate balance sheets, on both sides of the Atlantic. The cash buffers built up during the pandemic means we are unlikely to see the jumps in bankruptcies and in unemployment that we see in a more severe recession. How are these buffers looking currently?

Over the coming pages, we look closely at household and corporate balance sheets on both sides of the Atlantic to come to a judgment on whether financial buffers may blunt or delay the impact of credit tightening on economic activity. Overall, we judge that households are better placed to weather the coming recession than corporates, and that the US is better placed than the eurozone. However, buffers can only blunt or delay the onset of recessions, which we judge will be necessary to bring inflation sustainably back to target – both in the US and in the eurozone. As we stated in our March Monthly, should economies continue to prove more resilient than we expect, we think central banks will need to respond with further rate hikes in order to induce the necessary economic pain (falls in demand and rises in unemployment) to be sure that inflation falls fully back to 2%.

Box 1:

Credit and the economy (Aggie van Huisseling)

The tightening of credit standards and the impact on economic activity and thereby GDP growth works through two channels. Firstly, the tightening of standards means less credit availability and thereby less spending by firms or households that depend on banks for credit. Secondly, lenders may become cautious because a tightening of standards could indicate disturbance ahead. All in all, lending standards are a strong leading indicator for economic activity, with tighter standards implying reduced economic activity (link). Surveys on bank lending, such as the US Senior Loan Officer Survey or the ECB Bank Lending Survey (BLS) contain data on loan officer expectations on credit standards. As one would expect, these lead credit growth (link). Additionally, credit standards appear to function as an amplifier to shocks to the macroeconomy, as posed by Bernanke, Gertler, and Gilchrist (link). This means endogenous credit standards can operate as financial accelerator.

What does the literature tell us?

Various papers confirm the channels described above. Firstly, surveys on bank lending appear to be a strong leading indicator for credit growth. De Bondt et al. (2010) find that the ECB’s BLS responses significantly lead loan growth by one quarter for household lending and four quarters for enterprises. These results are consistent with those from the US Senior Loan Officer Survey (link). It also finds that – on top of interest rate and demand effects – the net tightening of credit standards based on BLS responses causes quarterly bank loan growth to decline by 1.3 percentage points. This finding is confirmed by Köhler-Ulbricht et al. (2016) which also find that BLS indicators lead bank lending growth and are well-anchored with financial developments (link). Secondly, credit standards can be viewed as early indicators of GDP growth (link, link, link). De Bondt et al. (2010) finds that BLS responses to a tightening in credit standards to enterprises leads real GDP growth by three to four quarters, with an ultimate impact on euro area real growth of between 0.8 and 1.0 percentage points during the 2008 financial crisis (link).

What do we see in the data (i.e. the historic correlations)?

To confirm the above findings, we looked at the historic correlations for GDP, business fixed investment, the credit impulse (for corporations, households, consumer and house purchase), credit demand (for firms, housing and consumer), and BLS credit standards (for firms, housing and consumer). The correlations between GDP growth and the listed indicators are strongest with the consumer credit impulse (0.41), the previous quarter net changes in housing credit standards (-0.45) and consumer credit standards (-0.46). This implies that tighter credit standards as indicated in the BLS – as well as the actual decline in credit – correlates with a decline in GDP. Next quarter responses to housing and consumer credit standards also correlate with GDP growth, but the impact is less strong (-0.34 and -0.35 respectively). The correlations with business fixed investment growth are similar to those described before for GDP; but now corporate credit also correlates more strongly. That is, for the corporate credit impulse we see a 0.39 correlation with business fixed investment growth. The correlations with the BLS net change in credit to firms last and next quarter and investment are -0.45 and -0.33, respectively. This shows that credit to firms more strongly correlates with investment than with GDP, which fits the intuition. Naturally, we observe strong negative correlations between the BLS variables and the credit impulses, which shows that tighter credit standards and lower credit impulse go hand-in-hand. Finally, credit demand positively correlates with GDP growth; but most strongly for consumer credit and housing credit demand.

Surveys have become less reliable since the pandemic, and so greater caution is warranted

Although bank surveys have been a strong leading indicator for economic output historically, we would interpret current survey results with greater caution in the current environment. Over the past year we have observed a decoupling of surveys such as PMIs and consumer confidence from economic activity, and it remains to be seen if bank lending surveys retain their explanatory power.

US: Household balance sheets have never been in better shape

Whichever way you look at it, household balance sheets in the US look very healthy. Household debt, the debt service ratio, and delinquencies are all historically low. The stock of consumer credit (including credit card debt) is still below the pre-pandemic trend. And the vast majority of mortgages are at fixed rates for the duration of the mortgage term (typically 30 years), insulating most households from rising mortgage rates.

Households also continue to hold significant excess savings from the pandemic period. According to the Fed’s flow of funds, checkable deposits totalled $4.8tn as of Q4 (18.3% of GDP), down only a little from the peak in Q3, and nearly five times the pre-pandemic level. As is customary in any discussion on household savings, we must add the caveat that the bulk of this is skewed towards higher income households, with the Fed’s distributional financial accounts suggesting that around 70% of the excess accrued since the pandemic sits with the top income quintile, with the bottom 40% holding just 10% of the excess. Meanwhile, the recent rise in the savings rate – albeit still holding at levels below the pre-pandemic norm – does suggest that households are approaching a limit in how willing they are to use excess savings (see chart on page 5). With lower income households having much smaller (in many cases zero) buffers, and given the continued pressure on real incomes from high inflation, we still therefore expect consumption to stagnate this year. But overall, it is clear that low household debt, combined with substantial cash buffers, will continue to cushion the expected downturn in consumption.

Eurozone: Household buffers have melted away

In the eurozone, total household net worth rose by over 7 trillion euros (66% of disposable income) over 2020-21. This enabled consumer spending to rise at a faster pace than disposable income over subsequent quarters, which is indeed what happened in 2022. Consequently, more than half of the extra net worth that had accumulated during the pandemic had melted away by the end of 2022. Importantly, the stock of liquid financial assets has almost returned to pre-pandemic levels, and only illiquid housing wealth remains significantly higher than before the pandemic. This has two consequences. First, eurozone households can no longer finance excess consumption by using extra accumulated liquid assets. Moreover, house prices have begun falling, and we see indications that households are now attempting to replenish total net worth by limiting consumption and saving more. Indeed, the household saving rate increased in Q4 22 for the first time since the first wave of the pandemic, and the savings rate is now above the pre-pandemic normal.

Impact of higher interest rates on consumption moderated by fixed rate mortgages

Recent research by the ECB (here and here) highlights that the impact of higher interest rates on household debt service ratios will be limited in the near term due to the high share of fixed-rate mortgages in many countries. Nevertheless, the share of flexible rate mortgages in new contracts has jumped recently, as households do not want to lock in current high rates for too long. The ECB reports suggest that consumers are already incorporating the impact of higher interest rates in their economic decisions, particularly in their plans for discretionary consumption. Indeed, this is has become visible in the incoming hard data for goods consumption, with retail sales volumes falling -1.0% 3m/3m in February, following a 0.6% qoq contraction in Q4 22 (see also the eurozone chapter).

With that said, as in the US, household debt is historically low in the eurozone (see chart on page 7). As such, although interest rate hikes are starting to bite, and consumers have clearly reached a limit in their use of excess savings, the strength of balance sheets will still help to prevent the coming downturn from becoming severe.

Households in the US appear more willing to use buffers than in the eurozone

One final point we must emphasise on the willingness of households to deploy buffers is: wealth preferences matter. The clearest indication of this is the savings rate, which historically is relatively stable. A savings rate below ‘normal’, pre-pandemic levels suggests a willingness to run down savings, while a savings rate above normal suggests a preference to rebuild buffers. In this regard, it appears that 1) households in the US have been much more willing to run down excess savings than those in the eurozone, and therefore 2) there is more downside risk to consumption in the eurozone than in the US. This supports our view of a more pronounced recession in the eurozone than in the US.

US corporate balance sheets not as strong as for households, but still healthy

On the corporate side, balance sheets are not nearly as strong in the US as for households, but fundamentals still look solid. Corporate debt is relatively high at around 100% of GDP, though given that much of this is fixed at relatively low rates from prior to the recent surge in rates, the debt service ratio is close to the long term average. Meanwhile, cash balances are off their pandemic highs, but still historically elevated, and the share of profits in national income is also historically high. As a cross-check for financial stress, bankruptcies fell during the pandemic period and remain exceptionally low.

All of this does not mean business activity has been immune to the tightening in financial conditions. On the contrary, business investment has contracted for the past three quarters, and we expect investment to remain weak for much of the year ahead (the Inflation Reduction Act will provide some support for investment, but more in the medium term). At the same time, while we do expect strength in balance sheets to prevent the kind of surge in bankruptcies we see in a typical recession, we still expect a rise in bankruptcies back to more normal levels. High interest rates are likely to put increasing pressure on balance sheets as debt comes due for refinancing – leading to a likely rise in debt service ratios – while cooling consumer demand will likely reduce pricing power and in turn put downward pressure on profit margins.

Eurozone corporate buffers could blunt credit tightening impact, but not avoid it

Non-financial corporations in the eurozone have accrued solid financial buffers since the pandemic. On the income side, profit margins have expanded on aggregate, which has helped lift cash balances to well above pre-pandemic levels. On the cost side, despite rising interest rates since H2 of last year, debt-service-ratios have so far remained stable and debt as share of GDP is in most countries on a declining trend since the pandemic induced rise.

The strength of balance sheets gives corporates a significant buffer against tighter credit conditions. Buffers can be used to deleverage in order to keep debt ratios low in a rising rates environment, kept as a rainy day fund to avoid bankruptcy as demand falls, or to help self-finance investment in an environment of reduced credit availability (although the latter is less likely given the weak growth outlook).

Buffers are more a source of short term relief rather than lasting strength. The blunting factors are temporary, which means tighter credit conditions will eventually have an impact. At the same time, as with household wealth, there are significant disparities beyond the aggregates. Size and sector differences matter, as was apparent with previous episodes where lending was constrained. During the sovereign debt crisis bank lending to SMEs was fell disproportionately compared to larger firms (see chart below). Even if SMEs have higher buffers now, they are probably less than the aggregates suggest, meaning tighter credit conditions will still have an impact. In a recent speech by ECB Chief Economist Philip Lane, a distinction was also made between ‘old’ and ‘young’ companies, with young companies (who generally lack buffers) facing much greater financing difficulties in an environment of tightening lending standards than more established companies.

All told, we judge that buffers help blunt or delay the impact of credit tightening, but that some significant impact on activity is inevitable. Indeed, as with the US, investment in the eurozone contracted in Q4 22 (see chart on page 5), and we expect tighter credit conditions to weigh on investment throughout the coming year. Bankruptcies have also surged, and are now well above the pre-pandemic level (see chart below). (Bill Diviney, Aline Schuiling, Aggie van Huisseling, Arjen van Dijkhuizen, Jan-Paul van de Kerke)

Box 2:

An ECB study (link) shows that the corporate debt composition also plays a crucial role in the impact of credit conditions. The use of bond financing relative to bank borrowing has expanded to 30% for euro area firms, up from around 15% in mid-2008. The study concludes that – when responding to an adverse aggregate supply shock – the option of bond finance blunts the impact of a tightening in the bank lending transmission channel. Working the other way, bond finance also amplifies monetary policy transmission channels that have broader financial market implications (such as QE). Countries with a low share of bond finance are more impacted by changes in short-term interest rates, while those with a high share of corporate bond financing are more affected by measures impacting the long-term rates.

Box 3:

Chinese lending is turning the corner after Zero-Covid exit (Arjen van Dijkhuizen)

While in developed economies credit growth is slowing on the back of aggressive rate hikes, the Chinese credit cycle is currently going in the opposite direction. The rapid exit from Zero-Covid, a more growth-oriented policy stance, and increasing signs of a stabilising property sector are the most important drivers. We think the turn in the credit cycle has further to run, although we do not expect a credit boom similar to for instance the one seen after the global financial crisis.

The property sector slump and Zero-Covid dampened credit growth in recent years…

In recent years, credit growth in China has been impacted by a number of factors. First, a structural impact comes from the financial deleveraging/derisking campaign initiated in late 2016. The goal of this campaign was to keep overall leverage in check by roughly matching lending growth with nominal GDP growth, while curtailing the most risky parts of the financial system such as shadow banking. Second was the introduction in August 2020 of the ‘three red lines’ policy for the highly leveraged property sector, which ultimately triggered payment distress among a wide range of property developers (including giant Evergrande), and culminated in a severe property sector downturn. Third, repeatedly (including in 2022), the pace of lending was impacted strongly by the combination of Covid-19 flare-ups and strict Zero-Covid policy resulting in recurrent broad lockdowns. On the household side, for instance, Zero-Covid and the problems in the property sector led to a sharp drop in consumer confidence and a collapse in demand for mortgage loans.

…but the credit cycle has turned after the end of Zero-Covid, with a more growth-oriented policy stance

Already in the course of last year, the Chinese authorities started with finetuning macroeconomic policy, and shifting the pendulum back to growth stabilisation, as the economy was hit hard by Zero-Covid and the property downturn. They resorted once more to piecemeal monetary easing, rather than introducing a GFC-style credit bazooka, by implementing mini cuts (of 15-35 bps) of several policy rates. The 5-year loan prime rate – used as a benchmark in mortgage lending – was cut the most (by 35bp, to 4.30%) to support real estate. Additional space for banks to lend was created by the further reduction in the reserve requirement ratio for banks, by 50bp in 2022, and a 25bp cut, to 10.5%, in March 2023. The authorities also used moral suasion to convince state banks to increase lending, and took specific steps to ease financing constraints for property developers. They also raised quota for local government bond financing to support infrastructure spending; the downside of this is that off-balance sheet borrowing through so-called local government financial vehicles (LGFVs) has risen to almost 50% of GDP, with LGFVs in poorer provinces experiencing payment difficulties.

Still, the most impactful policy shift was the abandonment of Zero-Covid initiated in December 2022. Though the messy character of this exit initially contributed to a further slowdown in lending, the disturbances from Zero-Covid exit proved rather short-lived. The Chinese economy has since clearly rebounded, led by the services sector, consumer confidence has started improving (although being still at relatively low levels), the housing market shows more and more signs of stabilisation, mortgage lending has started to pick up, and overall credit growth is accelerating again. This was illustrated for instance by the better than expected lending figures for March, and by the fact that the overall loan demand index in the PBoC’s banking survey jumped to a 9-year high of 78.4 in Q1-2023.

Turn in credit cycle has further to run; China’s rebound is cushioning global slowdown

Against this background, we assume the turn in China’s credit cycle has further to run, although we expect the authorities to continue to aim at keeping the overall leverage ratio in check, take measures to contain LGFV debt, and prevent the type of overheating issues that arose after China’s rapid rebound from the initial Covid-19 shock in 2020. On the global scale, we think that China’s reopening rebound and the turn in the credit cycle will help to cushion the impact of the slowdown in developed economies to some extent, though we do not think this will be a major game changer given the fall-out we expect (and are already experiencing) from unprecedented sharp rate hikes in the developed world, and with monetary policy working with famously long and variable legs.