Market Perspective - The butterfly's wings
Is Silicon Valley Bank (SVB) the butterfly which will cause a financial tornado?
After 2008, nobody should ignore it. A market-driven economy is always vulnerable to a seizure (and not just its financial sector). The collapse of even a small player can have consequences. But knowing that such an event could cause the next financial crisis is not the same as thinking that it will.
Certainly, SVB’s collapse – and in Europe, the directed takeover of Credit Suisse by UBS – has quickly refocused cyclical risk. In early March, investors were more worried about interest rates than earnings; suddenly, that mix has reversed. Contagion would bring a monetary tightening in excess of what even the biggest inflation hawk might advocate, and hit the economy and corporate profits hard. Such a deflationary shock would make tighter policy inappropriate, leading the Federal Reserve (Fed) and other central banks to reassess rates. They have already acted swiftly and significantly to keep the banking system liquid, even as their inflation targets continue to be missed.
Nonetheless, we do not see a wave of contagion ahead. We see lower inflation remaining the policymakers’ priority, and still think that can be achieved without a severe economic setback. The European Central Bank (ECB) at least has not yet been deterred from raising rates, even with Credit Suisse on its doorstep. Nor has the Fed, nor even the Bank of England (BoE).
We set out our thinking in the first essay below. In the second essay we suggest inflation is indeed continuing to moderate, albeit patchily.
Kevin Gardiner/Victor Balfour/Anthony Abrahamian
Global Investment Strategists
The butterfly's wings
A neat analogy...
The flapping of a butterfly’s wings causing a distant tornado is an idea from chaos theory that has become a popular description of How The World Really Works (Vaclav Smil, 2022). It says that a tiny, seemingly inconsequential event can have dramatic consequences as it cascades through a complex and non-linear system like weather formation – or the interdependent modern economy.
But there are vastly more butterflies than tornados. The proposed butterfly effect offers an analogy. If small events always had huge consequences, we would not be here to write and think about them. Complex systems do not always deliver extreme outcomes.
The Global Financial Crisis (GFC) which erupted with the collapse of Lehman Brothers on 15th September 2008 showed that a single institution can indeed cause global damage. But that came after a prolonged surge in bank lending and a widespread obfuscation of risk – sorry, financial innovation – which left bank and insurer balance sheets both stretched, and massively (and unpredictably) entangled. That single institution was systemically important, though we didn’t realise how much so at the time.
It took a while for the penny (or two) to drop. The authorities were impressively quick with their initial firefighting, but it took a while longer for more comprehensive support to be put in place (there was much wrangling over the US’s Troubled Asset Relief Program, for example). The VIX – the so-called ‘fear index’ which measures implied US stock market volatility – averaged more than 50% in the following six months, peaking at a sensational 80% (its reading on 22nd March 2023 was 22%).So the failure of a single financial business can have profound consequences (and Lehman was not the first such instance). Whether it does, however, may depend on the circumstances.
...But circumstances vary
Recently, the US bank system has if anything looked over-capitalised – lending growth in the last decade or so has been unremarkable, and banks have had equity to spare (leading them to buy back stock) and ample liquidity (figures 1–2 on page 4). Financial complexity has not surged as it did in the noughties, leaving the system more transparent and likely less sensitive to shocks. The response of the authorities has been fast and determined, with the US quickly offering to make good all deposits in the failed bank (even though this has revived the long-standing question of whether banks which are backstopped ought to be allowed to make more than utility-level profits).
SVB’s depositors (liabilities) and borrowers (assets) seem to have been highly concentrated in the riskier segment of the technology sector. Nervous tech entrepreneurs and employees tried to withdraw deposits just as the reduced value of their businesses was undermining the loans financed with those deposits. The bank is also reported to have been unusually exposed to directional interest rate risk (falling bond prices, a risk which is often hedged). A handful of other US banks have also faced difficulties, but have been met with promptly organised support.
Meanwhile, the Fed has continued to nudge its policy rate higher, but by a smaller amount (25 basis points) than had seemed likely (50 basis points) immediately before SVB’s collapse. At the same time it is making targeted liquidity available, and sounding less hawkish in tone.
In a different context, but influenced no doubt by the greater sensitivity to bank risk that followed SVB’s collapse, a bigger European bank, Credit Suisse, was also experiencing difficulties until its acquisition, at the direction of the Swiss authorities, by UBS on 19th March. Credit Suisse was experiencing outflows months ago: it has been serially unlucky in recent years, enmeshed in successive restructurings, and its stock price even before this latest stumble was less than a twentieth of its pre-GFC peak (and less than a fifth of its level five years ago). It has been labelled a systemically important institution, but its predicament and circumstances look idiosyncratic. If depositors in the new, better capitalised entity sit tight, Credit Suisse’s passing might even represent the removal of a potential source of instability.
European banks have recently been seen as more vulnerable than those in the US, and they may be more enmeshed, national boundaries notwithstanding – US banking remains regionally fragmented, which might offer some natural firebreak in case of localised emergency. But they are also collectively better capitalised, and less risky, than they were in 2007 (figures 1–2), and compare favourably on these metrics with their US peers (though local regulatory attitudes can still vary). While the ECB and indeed the Swiss National Bank have just raised interest rates a further 50 basis points, and the BoE by a further 25 basis points, European central banks are clearly aware of the risks, and are offering targeted liquidity and other support even as they nudge policy rates higher.
A less-fragile western banking sector has been accompanied of late by continuing low levels of unemployment, and – pandemic excepted – a macroeconomy that has (so far at least) been relatively well-behaved. Interest rates have risen sharply in the last year, and as noted, may not have finished rising yet, SVB and Credit Suisse-inspired nerves notwithstanding. But in real terms, they remain historically subdued, and there are few signs of economic trauma (again, so far).
Interest rates being previously so low for so long must have created some froth: they have surely contributed to the surge in inflation at least. But that froth is arguably less widespread, and less systemically important, than it could have been. Cryptocurrencies, NFTs, SPACs – these may be embarrassing, but may not do much damage (witness the earlier collapse of the FTX crypto exchange). It would have been remarkable for the normalisation of nominal interest rates to have passed completely without incident, and no doubt others lie ahead. But a full-blown crisis need not ensue.
Admittedly, if investors collectively worry enough about wider bank risk, their worries could become self-fulfilling, and the points made above might count for nothing. Market economies are vulnerable to shifting expectations, however sound or otherwise those expectations might be. If enough customers and peers believe that banks – or any other sector, for that matter – have a problem, then banks do indeed have a problem, and will be challenged. Investment advisers should guard against offering over-confident advice.
But our judgement currently is that SVB’s collapse, while unsettling, need not be – probably won’t be – the flapping butterfly wings that causes the next financial tornado. Contagion to date has been muted, and the slide in bank stocks needs to be seen in perspective (figures 3–6).
If the risk of a financial crisis has risen – albeit modestly – it would be surprising if the Fed and the other big central banks did not now rethink their appraisal of prospective inflation and monetary conditions, if only at the margin – the latest hikes notwithstanding.
A major loss of liquidity – a seizure of the banking system – is itself a tightening of monetary conditions, potentially on a massive and deflationary scale. If it happened, there would be no need to raise interest rates further: indeed, they might even need to fall.
In practical terms, in a seizure, the Fed and others might not be able to fine tune interest rates even if they still wanted to. Ultimately, monetary policy cannot be divorced from banking regulation and supervision, because if banks fail, the notion of incremental changes in the cost and quantity of liquidity, and of a neat ‘transmission mechanism’ from policy rates to the wider economy, goes out of the window.
In such circumstances there is effectively a ‘liquidity trap’ in the money markets. This may seem obvious, but it took the GFC to remind us of it. Ireland’s game-saving decision then to underwrite local bank deposits – a decision that arguably was not its to take – was perhaps the most visible illustration of what was at stake.
In the weeks ahead, then, we would expect the various central banks to be more restrained in their monetary hawkishness than they might otherwise have been (as noted, the Fed arguably already is). This will not mean that they have ‘gone soft’ (again) on inflation, but that they recognise that something other than higher interest rates could bring it down – and that bigger things might be in play.
The money markets have been quick to recognise the altered balance of risks. They have priced out much of the previously anticipated final stages of tightening, and priced back in some expected rate cuts before year end. These expected rates are however no more infallible than the earlier expectations – which indeed might yet turn out to have been more valid.
If we are right, and contagion is modest and a full-blown crisis is avoided, then we – and the Fed, and the ECB, BoE and Swiss National Bank – will return to watching the economic data unfold. As we write, the story the data have been telling is a clear one, and it feels too soon to change it materially.
That story so far: the global economy has slowed but not cracked under the weight of higher interest rates (perhaps for the reasons we outlined in the last Market Perspective). In Europe in particular another cyclical headwind – the worsened terms of trade caused by surging natural gas prices – is becoming a tailwind. The slowdown, together with improved supply conditions – including a gradual rebound in labour participation rates – has been allowing inflation to fall, albeit patchily and slowly (see below). In turn, this trend has been suggesting that the eventual peak in rates might not have been that far off – though if the economy were to stay resilient, an early reduction in rates would be unnecessary and unlikely.
Slowing but resilient output, declining inflation, and the prospect of an end to rising interest rates – this was not a bad economic prospect (we thought) for 2023. The ongoing strife in Ukraine, and (especially) the continuing tension over Taiwan, pose a testing geopolitical backdrop, but one which is now relatively familiar, and which may remain contained (at least in the narrow economic context). We felt that the big investment questions for 2023 would be when – and what – to buy, not whether to sell.
We had been waiting for what we saw as residual interest rate and earnings risks to be priced in more fully. In recent weeks – before SVB, and now Credit Suisse – it looked as if expectations for corporate profits might indeed be bottoming out, while money markets still displayed more of a peak profile than we had in mind (though that was beginning to change).
The resurfacing of bank risk has reversed that mix overnight. The increased possibility of a financial crisis means that corporate earnings are back under a cloud, while interest rate expectations (as noted) have fallen back (and taken on even more of a peak profile as they’ve done so).
We still feel that stocks, not bonds, offer the best prospects for long-term inflation-beating returns, though that revived earnings risk adds to our feeling that we should stay in ‘wait and see’ mode before advising a significant increase in holdings.
If banking risk subsides, as we think it will, the inflation story could likely remain as this year’s ‘hot topic’. Promisingly, headline inflation rates have continued to subside across most major economies, though core rates – that is, excluding food and energy – appear stickier (figures 7 and 8).
The inflation backdrop has improved most clearly in the US: the headline inflation rate has been declining since June, while core inflation has also been slowly moving lower. In fact, three of the four categories in the CPI basket have been improving:
- Energy. This component is largely shaped by US gasoline prices, which in turn are highly sensitive to changes in the price of crude oil: as the latter has declined, energy’s contribution to headline CPI inflation has also quickly diminished. Favourable base effects also mean that further declines in year-over-year energy CPI growth are possible even if oil price levels rise from here.
- Food. Its contribution to headline CPI inflation has only fallen slightly. That said, agricultural prices have been moving lower for almost a year according to the UN’s FAO Food Price Index, and so US food CPI inflation should abate further in the coming months (the price of the food on supermarket shelves also includes processing and distribution costs, which may explain why we haven’t yet seen a more pronounced decline).
- Goods. Supply chains appear to have normalised after China’s reopening, according to the New York Fed’s Global Supply Chain Pressure Index (figure 9) and goods-related inflation has cooled – even as US consumer spending has remained firm. The US producer price inflation rate has more than halved, suggesting the good news is percolating through the distribution chain: the cost pass-through from firms to consumers may be smaller going forward.
Services are the fourth – and stickiest – category. They account for almost 60% of the entire US CPI basket, with shelter – a gauge of housing and rental costs – accounting for the majority of that. Shelter CPI is yet to decelerate in year-on-year terms, but it tends to lag house price developments by roughly a year due to the way that it is measured (shelter CPI captures average costs across the entire housing stock, rather than the cost of recent transactions only). It will take time for recent, less expensive transactions to affect the average: US house price growth started to slow noticeably some months back, and so shelter inflation should also start to roll over at some point this year (figure 10).
The inflation backdrop has improved most clearly in the US: the headline inflation rate has been declining since June, while core inflation has also been slowly moving lower.
Europe’s inflation backdrop appears more concerning on first glance. The euro area headline inflation rate has only moved modestly lower and core inflation is yet to peak, while UK inflation rates turned higher again in February. The surge in natural gas prices has been a major contributor to headline and core inflation rates – the latter has been affected by the higher gas costs incurred by business have been passed on to consumers – and the past year’s US dollar strength may have also increased import costs. Swiss inflation, on the other hand, remains muted, perhaps due to its more stable domestic energy supply, stricter regulation and the stronger franc.
Even so, natural gas prices have now nearly fallen to mid-2021 levels, more than reversing the supply shock which arose from Russia’s invasion of Ukraine (figure 11). The impact of this collapse is unlikely to have been fully reflected even in headline European inflation rates yet. There are lags associated with the pass-through from wholesale gas prices to consumers (some estimates suggest that it can take up to six months), not least because of different governments’ energy support packages. This means that there is likely pent-up – but unreleased – energy deflation brewing in several European countries.
Meanwhile, despite tight labour markets and ongoing industrial unrest, real (inflation-adjusted) wage growth has remained negative in the US and Europe (figure 12). It’s possible that it may yet turn positive – if we used the Atlanta Fed’s US wage tracker in figure 12, for instance, it would be indicating (very) modest real wage growth – but most likely because headline inflation is falling: nominal wage growth appears to have already peaked in both the US and UK, and remains subdued in the euro area. All in all, a wage-price spiral seems increasingly unlikely to us (for context, real wages were rising for most of the 1970s despite double-digit inflation rates).
In summary, the global inflation backdrop is slowly improving, and will likely continue to do so over the course of this year, albeit gradually. And while, as noted above, we do not expect a widespread bank crisis, should one occur, that would likely reinforce this trend. Perhaps for the first time in many months it feels as if inflation risk is no longer tilted upwards.
Past performance is not a guide to future performance and nothing in this blog constitutes advice. Although the information and data herein are obtained from sources believed to be reliable, no representation or warranty, expressed or implied, is or will be made and, save in the case of fraud, no responsibility or liability is or will be accepted by Rothschild & Co Wealth Management UK Limited as to or in relation to the fairness, accuracy or completeness of this document or the information forming the basis of this document or for any reliance placed on this document by any person whatsoever. In particular, no representation or warranty is given as to the achievement or reasonableness of any future projections, targets, estimates or forecasts contained in this document. Furthermore, all opinions and data used in this document are subject to change without prior notice.