The forecast here at the start of the year was that the global manufacturing PMI new orders index – a key indicator of industrial momentum – would reach a peak in early 2021 and fall into the summer. The index declined slightly between November and February but rose to a new recovery high in March, with flash data last week and today’s Chinese results indicating a further significant increase in April. What has gone wrong?

The expectation of an early 2021 peak and subsequent relapse was based on a fall in global six-month real narrow money growth from an extreme peak in July 2020 – real money growth has led turning points in PMI new orders by 6-7 months on average historically. Six-month real narrow money momentum continued to weaken into March, so the monetary signal for PMI direction remains negative – see first chart.

Chart 1

There was meaningful variation around the 6-7 month historical average lead time. An April PMI new orders peak, were it to be confirmed, would imply a nine-month lead, which would be within one standard deviation of the average. So the further rise into April is not yet an unusual departure from the norm.

The most likely explanation is that the PMI upswing has been extended by US fiscal stimulus – particularly the third round of payments to households – along with initial moves towards economic reopening in the US, UK and other countries showing progress in virus containment. A 9.3% monthly surge in US retail sales in March may have been a key driver of stronger March / April new orders.

“Economic impact payments” authorised by the American Rescue Plan Act were $318 bn in March and $51 bn through 28 April for a total $369 bn, representing the bulk of a programme costed at $411 bn by the Congressional Budget Office.

New York Fed analysis of data collected in its monthly survey of consumer expectations indicates that households have spent or plan to spend 25% of the windfall, similar to the proportion in the first and second rounds, with remainder used to increase savings (42%) or pay down debt (34%). Rounding the $369 bn received to date up to $400 bn, this suggests additional consumer outlays of about $100 bn.

Assume that half of this amount is spent on goods, which could be an overestimate given that services account for two-thirds of total consumption. That would suggest additional retail sales – a rough proxy for goods spending – of about $50 bn. Monthly sales jumped by $47 bn between February and March. The suggestion is that the bulk of the boost to goods spending has already occurred and sales will fall back sharply into the summer.

Chart 2

An additional technical explanation for the March / April rise in PMI new orders is a positive base effect from the slump in the index to a low in April 2020. Survey respondents are asked to draw a comparison with the previous month but there is evidence that some replies take into account the level of business in the same month a year earlier – understandable in cases where there is a strong seasonal pattern in demand.

Specifically, a regression of the global manufacturing PMI new orders index on its one- and 12-month lagged values finds a small but statistically significant negative coefficient on the latter*. The coefficient suggests that a 13.7 point plunge in the index in March / April 2020 contributed 0.8 of a point to the estimated 3.0 point increase in March / April 2021 – third chart. This boost will reverse by June, reflecting the recovery in the index after April last year.

Chart 3

With global real narrow money growth still moderating, the US fiscal boost probably passing its maximum and China still on a slow growth path pending PBoC easing, the forecast here of a PMI pullback through late Q3 is maintained.

Chart 4

*The same result is obtained using US ISM manufacturing new orders data over a much longer sample.

The consensus has swung from pessimism about prospects for corporate profits in 2020 to likely excessive optimism now. Analysts, in particular, may underappreciate the contribution to recent profits resilience of government subsidies, withdrawal of which may offset much of the benefit of economic normalisation.

Posts here last year suggested that global profits would mirror V-shaped rebounds in retail sales and industrial output. S&P 500 aggregate operating earnings are likely to have risen above their pre-pandemic peak in Q1 2021, based on results to date and analyst estimates, according to S&P. Analyst forecasts imply growth of 29% between Q1 and Q4 2022 – see first chart.

Chart 1

S&P 500 operating earnings are equivalent to about 60% of corporate post-tax economic* profits in the national accounts and the two series follow a similar path. An attribution analysis, however, is available for the national accounts measure, allowing the recent contribution of higher subsidies and reduced taxes on production to be separated out.

As of Q4 2020, national accounts profits were 2.4% below their pre-pandemic peak in Q4 2019. If subsidies / production taxes had remained unchanged, the shortfall would have been 19.8%. Put differently, the rise in subsidies accounted for 17.8% of the level of profits in Q4 2020.

This direct contribution of government to profits will normalise as the economy reopens and emergency programmes are withdrawn. Assuming that the rise in subsidies since Q4 2019 is reversed, “underlying” profits – excluding the recent additional support – would need to grow by 21.7% from their Q4 2020 level to maintain overall stability. The consensus forecast of a 29% increase in S&P 500 operating earnings between Q1 2021 and Q4 2022, therefore, could require underlying growth of more than 50%.

Data later this week will show that GDP almost regained its pre-pandemic level in Q1 2021 and the FOMC’s median projection implies a further increase of more than 8% by Q4 2022. Adding in inflation at 2-3% pa suggests nominal GDP growth of about 13% – unlikely to support profits expansion of 50%.

Profit margins, indeed, could be squeezed by rising labour costs. The view here has been that the labour market would return to pre-pandemic levels of tightness by late 2021, resulting in upward pressure on wage growth. Labour market responses in the April Conference Board consumer survey support the view that conditions are normalising rapidly – second chart.

Chart 2

National accounts data for other G7 countries are less detailed / timely than for the US but government subsidies are likely to have provided similar or larger-scale support for profits. In the UK, the economy-wide gross operating surplus** would have been 13.0% lower in Q4 2020 without additional government subsidies / lower production taxes – third chart. The extra support amounted to 5.0% of GDP in Q4, double the equivalent in the US, reflecting the “generosity” of the UK’s furlough scheme.

Chart 3

*”Economic” = including inventory valuation and capital consumption adjustments.
**This is not comparable with the US national accounts profits measure – it is gross of depreciation, interest and income taxes and includes non-corporate business income.

Global six-month real narrow money growth appears to have edged lower in March, continuing a downtrend since last summer. This suggests that an expected relapse in global industrial momentum will extend through late Q3 / early Q4.

The global manufacturing PMI new orders index reached a new recovery high in March, consistent with a surge in six-month real narrow money growth into July / August 2020, allowing for the historical average 6-7 month lead time. More recent national surveys hint that March will mark a top – see chart 1.

Chart 1

The March real money growth estimate is based on information for the US, China, Japan, India and Brazil, together accounting for 70% of the G7 plus E7 aggregate tracked here. The US component is estimated from weekly data on currency in circulation and commercial bank deposits – official March money numbers are released next week and the Fed no longer provides weekly updates.

Global six-month real narrow money growth appears to have eased further to its lowest level since February 2020, reflecting stable nominal growth and another rise in six-month CPI inflation – charts 2 and 3.

Chart 2

Chart 3

Chart 4 shows the early reporting countries individually. US six-month real narrow money growth is estimated to have edged lower despite disbursement of $318 bn of stimulus payments to households – these were made in the second half of the month and may have a larger impact in April (the money numbers are month averages).

Chart 4

Six-month growth also eased slightly further in Japan and China, with a small rise in India. Brazil moved into contraction although this needs to be placed in the context of an extraordinary surge last summer – 12-month growth is still strong.

Markets could be starting to offer corroboration of the scenario of a global manufacturing PMI new orders peak and pull-back, with Treasury yields stalling and equity market cyclical sectors no longer outperforming – chart 5.

Chart 5

Will global six-month real narrow money growth recover? Six-month CPI inflation is likely to rise slightly further in April / May but could fall back in H2 as commodity prices move sideways or correct.

Fiscal stimulus is acting to push up US nominal money growth but there may be an offsetting drag across the G7 from recent bond yield rises – chart 6.

Chart 6

A revival in Chinese narrow money growth probably requires a PBoC policy shift. The view here has been that policy was overtightened in H2 2020 and the economy would slow in H1 2021. This scenario appears to be playing out, with Q1 GDP disappointing and industrial output falling in March. Core CPI inflation (i.e. ex. food and energy) is at 0.3% and a surge in PPI inflation reflects input cost rises that are squeezing downstream margins. The PBoC has allowed three-month SHIBOR to drift back to its January low, consistent with a switch to an easing bias – chart 7.

Chart 7

The assessment in the previous quarterly commentary was that the monetary backdrop for markets had deteriorated at end-2020. This was arguably reflected in weak bond market performance during Q1 but global equities rose further as earnings expectations were revised higher. The monetary indicators followed here continue to give a cautionary message for markets while suggesting that global industrial momentum will slow into late Q3. A summer growth “scare” could trigger a correction in equities and a recovery in defensive sectors.

The market assessment relies on two indicators of “excess” money, which, according to the “monetarist” view, is a key influence on demand for financial assets: the difference between global six-month real narrow money and industrial output growth, and the deviation of 12-month real money growth from a long-term moving average. The entire outperformance of global equities relative to US dollar cash since 1970 occurred during periods when both indicators were positive. Equities underperformed cash on average when either or both were negative.

Allowing for data publication lags, the indicators gave a joint positive signal at end-April 2020. The MSCI All Country World Index (ACWI) returned 33.9% in US dollar terms between then and end-2020, reflecting both a recovery in earnings expectations and a rerating of markets. The “buy” signal, however, was rescinded at end-December following a cross-over of real narrow money growth beneath industrial output growth – see chart 1.

Chart 1

Global equities derated during Q1 – the ACWI 12-month-forward PE ratio fell by 4.2% – but the index nevertheless returned 4.7% as forecast earnings rose by a further 8.6%. Earnings optimism was boosted by confirmation of additional large-scale US fiscal stimulus, which also contributed to continued outperformance of cyclical sectors. The view here, however, is that global industrial momentum is peaking and will slow through late 2021. This would be a shock to the consensus and could trigger an unravelling of recent market trends.

The slowdown forecast rests on the relapse of global six-month real narrow money growth from its July-August 2020 peak – turning points have led those in industrial output growth by nine months on average historically. The lead time on the global manufacturing purchasing managers’ index (PMI) is slightly shorter, suggesting that a new high in the index reached in March will mark the peak of the current upswing – chart 2.

Chart 2

China’s industrial recovery has already decelerated, with the Markit / Caixin manufacturing PMI falling to an 11-month low in March. Chinese monetary policy was less stimulative than elsewhere in H1 2020 and retightened in H2, explaining relatively weak money trends – chart 3. China’s PMI has led the global measure since the GFC – chart 4.

Chart 3

Chart 4

Global six-month real narrow money growth continued to subside in February. A recovery could unfold into the summer as US money numbers are boosted by disbursement of fiscal stimulus and if the PBoC relaxes policy in response to softer economic data. Such a scenario could result in another “excess” money buy signal by mid-year while suggesting industrial reacceleration from late 2021. A money growth rebound, however, is likely rather than guaranteed and the judgement here is that the focus for now should be on downside economic / market risks.

An industrial slowdown could be offset in GDP terms by services strength if covid developments allow economic reopening. This could, however, contribute to industrial deceleration by reversing last year’s substitution by consumers of goods for services spending. Industrial trends are likely to be more important for markets, partly reflecting a stronger correlation with equity earnings. Services-driven GDP strength could make central banks less inclined to offer support in the event of industrial / market weakness.

Global CPI inflation rates are spiking higher in reflection of recent commodity price moves and base effects but inflation worries could be near a short-term peak if the above industrial scenario unfolds – another reason for doubting that the cyclical / value rally will extend in Q2. Input price components of business surveys will fall away into the summer barring another surge in oil and other industrial commodities – chart 5. Cyclical sectors may be fully discounting “reflation”, judging by valuation relative to defensive sectors – chart 6.

Chart 5

Chart 6

The March rise in the global manufacturing PMI was driven by European components, with the US PMI little changed and China’s – as noted – easing further. Eurozone strength is consistent with a real money growth spike last summer but a subsequent slowdown argues against current levels being sustained – chart 7.

Chart 7

UK money trends, by contrast, are diverging positively from other majors, signalling a relatively bright economic outlook and possible support for UK equities – chart 3. Money growth strength reflects larger-scale monetary deficit financing than in other countries, which may continue given PM Johnson’s big spender bias and a supine Bank of England. “Excess” money could partly flow overseas, suggesting downside risk for sterling, in which speculators appear to have accumulated a significant long position.

The forecasting approach here uses cycle analysis as a cross-check of the monetary analysis and to provide longer-term perspective. The previous assessment, which is maintained, was that the stockbuilding and business investment cycles bottomed in Q2 2020, while the long-term housing cycle remains in an upswing. The suggestion that all three cycles were turning supportive for the global economy and markets reinforced the positive message from monetary trends in mid-2020.

The next scheduled low is in the short-term stockbuilding cycle. Based on an average historical cycle length of 3.5 years, this could occur in late 2023, with the downswing into the low starting 12-18 months earlier, i.e. in mid- to late 2022. Risk markets tend to weaken in the 18 months leading up to a cycle trough – major equity bear markets have usually occurred during this time window.

The suggestion is that the primary trend in the global economy and markets will remain up through H1 2022. Stockbuilding cycle upswings, however, typically unfold in a zig-zag pattern, with an initial upthrust followed by a corrective phase before a final move higher into the peak. The judgement here is that the initial phase is ending and cycle momentum will diminish into H2, consistent with the monetary forecast of an industrial slowdown. The view that the initial phase is mature is supported by the business survey inventories indicator in chart 8.

Chart 8

Market moves since Q2 last year, moreover, have in most cases matched or exceeded averages during 18-month periods following previous stockbuilding cycle lows – see table 1. Developed market equities, cyclical sectors and commodity prices, in particular, have performed strongly, suggesting limited further upside even though a stockbuilding cycle downswing may be a year or more away.

Table 1

Source: Refinitiv Datastream, own calculations

A further consideration is that the current stockbuilding cycle could be shorter than average. The covid shock appears to have extended the previous cycle to 4.25 years. If the current cycle were to display an offsetting deviation from the average 3.5 years, the next low would occur in early rather than late 2023 (i.e. 2.75 years from the Q2 2020 trough). This, in turn, would imply that the 18-month negative period for markets ahead of the low would start in H2 2021.

The latter possibility, it should be emphasised, is not the central case here and would require confirmation from a further fall in global six-month real narrow money growth during H1 2021 rather than the US-led rebound suggested earlier.

The comparison of recent returns with stockbuilding upswing averages, as well as supporting the case for reducing cyclical sector exposure in favour of defensive sectors, suggests relative value in emerging market equities, quality and gold, and scope for a further rally in the US dollar. Stronger EM equity performance, however, may be conditional on a recovery in Chinese money growth, probably requiring a prior PBoC policy shift.