US money growth, on a range of measures, picked up significantly during H2 2019. This suggests a revival of economic momentum in mid-2020 but near-term data could be surprisingly weak. The monetary acceleration, moreover, was unusual and could be in the process of reversing equally sharply.

The rise in money growth was not due to bank lending. Three-month growth of commercial bank loans and leases moderated through 2019, reflecting a shift from strength to weakness in commercial and industrial (C&I) loans – see first chart.

As previously discussed, C&I loan demand is correlated with the stockbuilding (inventory) cycle. The year-on-year C&I loan “impulse” at end-December, i.e. the change in three-month momentum from a year earlier, was the most negative since 2009, suggesting a large stockbuilding drag on annual GDP growth – second chart.

The rise in money growth was not primarily due to the Fed’s balance sheet expansion, which started in September. The Fed initially injected reserves via repo operations, which, when conducted with banks, have no direct impact on the stock of (broad) money. Treasury bill purchases have an impact only to the extent that they are purchased from non-banks.

The key driver of monetary acceleration, instead, appears to have been strong buying of Treasuries by banks. Broad money growth, as measured by the quarterly non-financial M3 aggregate preferred here, has shown an unusually high correlation in recent years with such purchases, which reached a dollar record in Q3 – third chart.

Banks stepped up Treasury buying to offset an excessive fall in their holdings of high quality liquid assets caused by the reserves drain from the Fed’s QT operations – fourth chart.

The Fed’s subsequent liquidity injection has boosted reserves and removed banks’ incentive to buy Treasuries. Weekly data on commercial banks’ assets suggest that purchases have dried up – fifth chart. (The weekly numbers are only a rough guide as they incorporate valuation effects.)

This cessation is being reflected in monetary data: six-month growth rates of weekly narrow and broad money measures have moderated since November – sixth chart. Further weakness would temper the positive signal from the earlier pick-up and suggest a renewed slowdown in economic momentum later in 2020.

UK rate cut sceptics argue that a rise of 135.000, or 0.5%, in the number of employees in the three months to November from the prior three months signals a strong labour market and will give MPC doves a reason to climb down.

This increase, however, merely reverses a decline earlier in 2019 – the November three-month level was unchanged from February. The stock of vacancies recovered slightly in the three months to December, which appears to reflect a sharp jump in December alone, but the signal for employment prospects remains negative – see first chart.

MPC members may be more impressed by confirmation that private sector earnings growth has peaked – second chart.

The labour force survey measure of unemployment has moved sideways since early 2019 but the “experimental” alternative claimant count, which attempts to track the number of jobless benefit recipients by adjusting for the roll-out of universal credit, continued to rise into November – third chart.

Today’s public finances release for December provides further evidence of significant labour market cooling, with pay-as-you-earn tax receipts, smoothed over six months, showing the slowest year-on-year growth since 2013 – fourth chart.

A fall in Chinese money growth during H2 2019, especially in real terms, suggests that the economy will remain weak through H1 2020. Previous posts recommended monitoring the Cheung Kong Graduate School of Business monthly survey of private sector firms for signs of an easing in credit and monetary conditions in early 2020.

The January results were released today and are hopeful: the corporate financing index, measuring the ease of obtaining external finance, rebounded sharply from December’s 11-month low – see chart.

There are caveats. The index remains far below its level in March / April last year – those readings proved a false recovery signal. The improvement may partly reflect a knee-jerk positive response to the US-China trade truce – survey components measuring expectations for corporate sales and profits showed much larger increases last month. A possible negative effect from the Wuhan coronavirus has yet to be incorporated.

January money and credit data released in mid-February will be the next test, though could be clouded by the early lunar new year. Confirmation of an easing of monetary conditions would have limited implications for H1 economic prospects but would give grounds for optimism about H2.

Six-month growth of global (i.e. G7 plus E7) real narrow money is estimated to have been unchanged at 2.4% (not annualised) in December, based on monetary data covering 70% of the aggregate and near-complete CPI inflation data – see first chart.

Real money growth remains well below a post-GFC average of 3.2%, suggesting that six-month industrial output momentum will show limited recovery through Q3 2020, allowing for an average nine-month lead.

Real money growth bottomed at 1.0% in November 2018 while industrial output momentum appears to have reached a low 10 months later in September 2019.

The recent high for real money growth was 2.6% in September 2019. A rise through this level, ideally to 3-4%, in early 2020 would signal economic acceleration during H2 and into 2021. This is the scenario suggested by cycle analysis – the stockbuilding and business investment cycles are expected to have entered recovery phases by H2.

A relapse in real money growth without a prior breach of the September high, however, would signal a “double dip” starting around mid-2020.

US trends could drive a relapse. Six-month narrow money growth surged higher in September as the Fed cut rates for the second time in two months and restarted balance sheet expansion – second chart. Rates have been stable since end-October and the balance sheet boost could be ending. Narrow money contracted in the first two weeks of January – third chart.

A fading impact of the ECB’s September policy easing could similarly result in a moderation of Euroland money growth. The Q4 bank lending survey released today signalled a further weakening of business credit demand – fourth chart. (December money numbers will be released on 29 January.)

An alternative positive scenario is that US / Euroland money trends remain solid while a reversal of recent Chinese real money contraction pushes the global six-month growth measure through 3%. A sharp fall in CPI inflation as pork prices normalise will give a significant boost to Chinese real money but the timing is uncertain, while – as previously discussed – a recovery in nominal money growth probably depends on an unwind of bank credit tightening during 2019, of which there is little sign yet.

The Chinese GDP numbers released on Friday appeared to show a strong pick-up in nominal growth in Q4. This was puzzling but possibly attributable to the recent food-driven inflation surge. Some analysts latched onto the pick-up as a bullish signal.

The numbers carried by Datastream changed over the weekend and now show a slowdown rather than acceleration in Q4. The explanation is that there have been upward revisions to the level of nominal GDP in previous quarters, which weren’t available on Friday – we were comparing the new Q4 number with old-basis data for earlier quarters.

The expectation here has been that two-quarter nominal GDP expansion would fall in H1 2020 in lagged reflection of declines in six-month growth rates of money and credit since Q2 2019. The new numbers suggest that that this slowdown has already started – see chart (revised from Friday’s post).