The global economy has rebounded strongly as virus control measures have been eased, with industrial output on course to exceed its pre-covid level in Q4 2020. Major central banks have ignored this positive surprise, locking themselves into multi-year maintenance of ultra-loose policies. Fiscal authorities, similarly, are fearful of winding down stimulus. The global broad money stock is far above its estimated “equilibrium” level, implying medium-term upward pressure on asset prices, economic growth and goods and services inflation.
Economic and market commentary remains focused on virus developments to the neglect of arguably more significant monetary and cyclical drivers. Global new cases were easing off in early October, while the European “second wave” – exaggerated in the data by increased testing – appeared to be peaking. The discussion below assumes that virus disruption will continue to wane, reflecting better targeted control measures, medical advances and a falling susceptible share of the population.
The monetary policy response to the crisis may have larger medium-term economic effects than the virus itself. An unprecedented surge in the G7 broad money stock in Q2 opened up a record 17 percentage point deviation with the estimated demand to hold money, with the gap remaining huge in Q3 despite the strong economic bounceback – see first chart. The money demand estimate is derived from a model in which desired holdings depend on nominal GDP, gross financial / housing wealth and real interest rates. This model explains the long-run downward trend in broad money velocity as a function of a rising wealth to income ratio and falling real yields.
The “monetarist” view is that the monetary imbalance will be eliminated over the medium term by some combination of a money supply reversal, strong economic growth, higher asset prices and a rise in goods and services inflation. A similar overshoot during the GFC was corrected mainly by a bank lending-driven contraction in the money stock and a strong rebound in asset prices from historically low valuation levels. Monetary financing of fiscal deficits and official support for bank lending argue against money supply weakness now, while equity, bond and housing valuations are much higher. “Excess” money, therefore, is more likely to be reflected in strong economic growth and a sustained inflation rise.
Broad money trends are informative about medium-term prospects for nominal incomes and wealth but narrow money is a better guide to short-term swings in economic momentum. Six-month growth in real (i.e. inflation-adjusted) narrow money in the G7 economies and seven large emerging economies eased in August from a July record high and will slow significantly further as bumper monthly rises over March-May drop from the calculation. Allowing for an average nine-month lead, this suggests a moderation of economic strength from Q2 2021.
Monthly growth in real narrow money, however, stabilised at a solid level in July / August, with US weekly data suggesting a September pick-up there. Narrow money trends, therefore, are consistent with an extension of robust economic expansion into mid-2021, at least.
Six-month real narrow money growth remains much higher in the US than elsewhere, reflecting larger-scale QE and rate cuts – third chart. Upside risks to economic growth and inflation, accordingly, may be greatest in the US. China’s policy response to the crisis was conventional and real money growth is lagging, although survey evidence of easing credit conditions suggests a further pick-up.
The monetary forecast of strong global economic growth through Q2 2021 and probably well beyond is supported by cycle analysis. Three investment cycles drive economic fluctuations: the 3-5 year Kitchin stockbuilding or inventory cycle, 7-11 year Juglar business investment cycle and 15-25 year Kuznets housing cycle. Consumer spending, while the largest demand component, is the least volatile and echoes the Juglar and Kuznets cycles, which influence employment trends and housing wealth respectively.
The 2018-19 global economic slowdown reflected joint downswings in the Kitchin and Juglar cycles coupled with an interest rate-driven pause in a Kuznets cycle upswing. The previous quarterly commentary argued that the covid shock completed the Kitchin and Juglar downswings, with both cycles likely to have reached bottom in Q2. Q3 data indicating an end to inventory depletion and a strong rebound in capital spending support this thesis, while low long-term interest rates have reenergised the Kuznets housing upswing.
The upshot is that all three cycles are now acting to push up global economic growth, with their combined impact likely to outweigh any fading of fiscal / monetary stimulus in 2021-22.
The fourth chart illustrates the global inventory cycle using G7 national accounts data and a more timely business survey indicator, which is giving strong confirmation of a Q2 low. Cycle upswings usually last at least two years, with an initial upthrust giving way to an interim correction before a final push into the peak. The judgement here is that the initial move higher has started and may carry into Q2 2021, which would dovetail with the narrow money signal of a moderation of economic momentum around then.
The stockbuilding cycle could also play a key role in validating the monetary forecast of a significant rise in global goods and services inflation in 2021-22. Cycle upswings are usually associated with upward pressure on industrial commodity prices, as illustrated by the fifth chart. The highest peaks in annual commodity price inflation since WW2 occurred in the upswings from cycle troughs in 1971 and 2009. Global annual broad money growth had reached multi-decade highs before these surges, and is above those levels now.
The policy-maker and market consensus assumes a disinflationary impulse from prolonged labour market slack but unemployment rates could return to pre-covid levels in H2 2021 in the strong growth scenario outlined above. US unemployment has fallen surprisingly quickly as workers on temporary layoff have been recalled and the “permanent” rate – excluding such layoffs – is at an average level by historical standards (5.0% in September versus 5.2% over 1985-2019). Use of short-time work subsidy schemes in Eurozone countries, similarly, has fallen sharply, suggesting that still-low official unemployment will rise by less than expected. A surge in US / UK business formations offers hope that new jobs are being created to offset covid-driven losses.
Current high global money balances suggest rising asset prices over the medium term but short-term market movements reflect the direction of “excess” money rather than its level. The forecasting approach here focuses on the differential between six-month rates of change of global real narrow money and industrial output – equities and other cyclical assets have outperformed cash significantly on average historically when this has been positive. Data on the differential are released with a two-month lag and allowing for this delay improves backtested results.
The real money / output growth gap reached a record high in April but may turn negative in October as base effects cause a temporary overshoot of industrial output momentum. If confirmed, and allowing for the reporting lag, this would suggest shifting defensively around end-2020.
Current still-solid monthly real narrow money growth, however, implies that a positive six-month differential could be restored in early 2021 as output momentum normalises. The historical behaviour of markets during Kitchin cycle upswings also argues that any weakness in cyclical assets would probably be temporary. The table compares returns in the current cycle, for which there is now strong evidence of a Q2 trough, with an average of seven prior cycles (returns are measured from the end of the middle month of the cycle trough quarter). The pattern of returns into and since the latest trough is consistent with historical experience, with individual comparisons in most cases suggesting scope for recent moves to extend over the next 12+ months.
