Japanese monetary trends continue to signal respectable economic performance but are less strong than in the Eurozone and US, despite record QE.
Narrow money M1 and broad money M3 rose by 0.3% and 0.2% respectively in March. Six-month growth rates eased to 2.6% for M1 and 1.6% for M3, or 5.3% and 3.2% annualised respectively – see first chart. For comparison, Eurozone M1 and M3 rose by 11.8% and 5.1% annualised in the six months to February, the latest available month.
M1 and M3 slowed in late 2013 / early 2014 – first chart. With inflation boosted by the April 2014 sales tax hike, real money contracted, correctly signalling economic weakness – second chart. Nominal money growth recovered during the second half of 2014 while six-month inflation has fallen back, reflecting both a reversal of the sales tax effect and weaker energy prices. Real money trends, therefore, are again consistent with respectable economic expansion.
The bigger story is that QE appears to have had little impact on monetary growth in recent years. The annual increase in M3 stood at 2.1% when the Bank of Japan (BoJ) launched QE in October 2010. It was 3.0% in March 2015. The BoJ is currently buying securities at an annual pace of ¥83 trillion, equivalent to 6.8% of M3. This suggests QE “pass-through” of 13% (i.e. 0.9 as a percentage of 6.8). An earlier analysis of UK QE arrived at a similar pass-through estimate of 21% – much lower than the Bank of England’s claimed 59%.
What explains the high “leakage”? An analysis of the monetary counterparts shows that the increase in BoJ lending to the government has been offset by a reduction in bank exposure. Banks have cut their lending because QE has boosted their reserves at the BoJ, meaning that they need to hold fewer government bonds to meet liquidity targets*. When QE started in October 2010, combined lending to the government by the BoJ and banks was contributing 3.3 percentage points (pp) to annual broad money growth, comprising 0.5 pp from the BoJ and 2.7 pp from banks** – third chart. On the latest figures, for January 2015, the combined contribution was 2.9 pp, with the BoJ adding 5.3 pp but banks subtracting 2.4 pp.
Near-term economic prospects may be improving but the BoJ deserves little credit.
*This response was predicted in a post in April 2013.
**Difference due to rounding.
A post in February suggested that global growth would pull back into mid-2015 before strengthening significantly in the second half of the year. This scenario remains on track judging from monetary trends and leading indicators.
Global growth is proxied here by the six-month rate of change of industrial output in the G7 and seven large emerging economies*. This rebounded between August and December 2014, stabilising in early 2015 – see first chart.
The OECD today released February data for its country leading indicators, allowing an update of the global longer leading growth measure followed here. This had been stable at a respectable level through January but fell back in February, consistent with a near-term loss of economic momentum**.
Monetary trends suggest that this pull-back will be modest and temporary. The real or inflation-adjusted money supply leads activity by six to 12 months, according to the monetarist rule. Global six-month real narrow money growth fell between March and August 2014, signalling slower economic momentum in early 2015. The decline was due to the US, where economic news has been surprising negatively, as forecast in a post in September.
Global real money expansion, however, rebounded strongly in late 2014, reaching a 38-month high in February. The global economy, therefore, should regain speed this summer and is likely to be growing at a rapid pace by late 2015. Again, the momentum change should be driven by the US, reflecting recent better monetary trends.
Second-half economic strength should be accompanied by a rebound in inflation, posing a risk to government bond markets. The six-month change in global consumer prices should stage a V-shaped recovery if oil and other commodity prices stabilise at current levels – second chart. This would probably pull down real money growth, in turn suggesting another economic slowdown in 2016. A fall in the gap between real money growth and output expansion, meanwhile, could imply less favourable liquidity conditions for equities later in 2015.
*Industrial output is preferred to GDP because it is more timely, less revised, available monthly and better correlated with equity market earnings.
**The February decline may have been exaggerated by the impact of bad weather and a ports strike in the US, and a late new year holiday in China.
The Bank of England’s preferred broad money measure, M4 excluding intermediate other financial corporations, rose by only 0.1% in February, following an identical January gain. This sluggishness probably reflects savers switching from bank deposits into National Savings pensioner bonds* and does not signal deteriorating economic prospects. National Savings attracted £3.2 billion in February following £6.9 billion in January, versus an average monthly inflow in 2014 of £500 million. This suggests that about £9 billion of the bonds were sold in January / February, equivalent to 0.5% of the stock of broad money.
National Savings competition also explains weak retail sales of unit trusts and OEICs so far this year – £0.8 billion last month and £0.3 billion in January versus a monthly average of £1.7 billion in 2014.
Narrow money has been less affected by the pensioner bond sales. Non-financial M1, comprising physical cash and sight deposits of households and private non-financial corporations, rose by 0.4% in February after a 0.5% January increase. Six-month growth, however, has fallen significantly since 2013 and is well below the current Eurozone pace – see first chart.
This slowdown, fortunately, has been largely matched by a (temporary) fall in consumer price momentum – second chart. Real narrow money growth, therefore, remains solid and has rebounded recently, suggesting respectable economic expansion through 2015 – third chart.
In other UK news, GDP growth in 2014 has been revised up from 2.6% to 2.8%, or 2.9% excluding the North Sea. Meanwhile, aggregate wages and salaries – i.e. the product of average pay and the number of employee jobs – rose by 5.3% in the year to the fourth quarter, the largest annual gain since a 5.8% increase in the fourth quarter of 2007. Adjusted for consumer price inflation (0.9% in the fourth quarter), growth was the strongest since June 2001, when the Blair government was reelected with another landslide majority.
*In terms of the monetary counterparts (i.e. the arithmetic linking changes in broad money to other elements of the banking system’s balance sheet), the bond sales contributed to public sector “overfunding” and a consequent repayment of bank borrowing.
Posts here from the summer onwards argued that faster Eurozone monetary growth, particularly of the narrow M1 measure, was signalling notably better economic performance in early 2015. The monetary signal was starkly at odds with consensus gloom, including recession warnings from the IMF and others, but has been vindicated by recent data. The Citigroup Eurozone economic surprise index has been strongly positive since early February, while the composite PMI output index rose to a 46-month high in March, according to this week’s flash data.
The earlier posts noted that narrow money growth was particularly strong in Spain, where the central bank now estimates that GDP will rise by 0.8% in the current quarter, following a 0.7% fourth-quarter gain.
The positive trends in M1 and the broader M3 measure continued in February, with media reports of the data containing numerous quotes from commentators about the bullish implications for economic prospects. Where were these experts six months ago?
M1 and M3 rose by 0.9% and 0.6% respectively in February alone. Six-month M1 growth increased to 5.7%, or 11.8% annualised – the highest since November 2009; the corresponding M3 measure was 2.5%, or 5.0% annualised – see first chart.
One reason for the change in the consensus interpretation of the data is that private sector credit is now expanding. Adjusted for sales and securitisations, bank loans to the private sector grew by 0.5%, or 2.2% annualised, in the latest three months. As noted with monotonous regularity in previous posts, the historical evidence for the Eurozone and elsewhere shows that money tends to lead the economy while credit is coincident or lagging. The credit pick-up, nevertheless, is significant because it confirms the positive message from other coincident data and will support broad money growth going forward.
The second chart shows the drivers of six-month M3 growth, based on the ECB’s monetary counterparts analysis. As well as the turnaround in private sector credit, M3 expansion has been pushed higher by bank buying of government securities, encouraged by the ECB’s TLTROs, and a switch in bank funding from bonds and other longer-term liabilities to deposits, reflecting lower interest rates. By contrast, growth in banks’ net external assets, which was a major supportive factor in 2013-14, has slowed sharply. With the current account surplus remaining high, this implies that Eurozone non-banks have been exporting capital – the ECB’s monetary boost, in other words, has partly spilled abroad.
A key issue is whether QE leads to further monetary acceleration. The assessment here is that QE programmes had limited impact on broad money trends in the US, UK and Japan. Annual growth in Japanese M3, for example, was 2.9% in February versus 2.5% in March 2013, just before newly-installed Bank of Japan Governor Kuroda launched his monetary blitz. The small impact of these programmes is judged to be for two reasons. First, central bank purchases of government securities were partly offset by sales (or reduced buying) by banks, whose demand for liquid securities fell as QE expanded their reserves. The effect on broad money depends on the combined transactions of the central bank and banks. Secondly, the payment of interest on reserves meant that banks were content to “hoard” their increased balances rather than use them to expand their assets. Asset expansion would not have reduced the total amount of reserves but would have increased their velocity of circulation.
Eurozone QE is also likely to involve offsetting transactions by banks; they appear, indeed, to have started already, with banks selling €25 billion of government securities in February, following the ECB’s QE announcement on 22 January. The second-round effect of higher reserves on bank balance sheets, however, could be larger than in the US, UK and Japan because banks are penalised for holding excess balances via the negative (-0.2%) deposit rate. The expansion of reserves, in other words, is less likely to be fully offset by a fall in their velocity.
Current Eurozone money supply trends signal that monetary conditions are already sufficiently, and possibly excessively, loose. A further acceleration in response to QE would suggest that the ECB will be forced to call time on the programme well before the indicated September 2016 termination date.
The annual change in consumer prices fell to zero in February, prompting media headlines about “record low” inflation. A more correct statement is that CPI inflation is at a 25-year low, since official statistics do not extend back before 1989.
For a longer-term perspective, it is necessary to use the previously-targeted RPIX (retail prices excluding mortgage interest) measure*. The annual change in RPIX was 1.0% in February, matching a low reached in June 2009. Looking further back, RPIX inflation was below the current level in 1963, 1960, 1959 and 1954. So “record low” is incorrect even confining attention to post-WW2 history.
The current deviation of the annual CPI change from the 2% target is explicable by energy and food price weakness and sterling strength. “Core” inflation excluding energy, food, alcohol and tobacco was 1.2% in February, while sterling’s appreciation has probably lowered the core rate by at least 0.5 percentage points.
Services prices are a better guide to domestic inflationary pressures, since they are less affected by changes in commodity prices and the exchange rate (though not impervious**). Services inflation was stable at 2.4% in February versus a 2014 average of 2.5%.
Solid monetary trends continue to suggest that the next big move in domestically-generated inflation will be higher. The headline rate may rebound surprisingly strongly later in 2015 and in 2016 as the commodities / sterling drag unwinds.
*A recent paper by a former government economist disputes the official view that the CPI is statistically superior to the RPI / RPIX.
**Examples of effects include food costs on catering services, energy costs on transport services and the exchange rate on foreign holidays.