China is stepping up policy easing, as warranted by recent weak monetary and economic data – see previous post. The three-month repo rate this week fell to its lowest level since March 2011, almost converging with the officially-set one-year deposit rate, of 3.5% – see first chart.

The easing process should be facilitated by better inflation news. Chinese food prices usually fall into mid-year but the decline in 2012 has been larger than in 2011, judging from weekly statistics – second chart. CPI food inflation may subside from an annual 7.0% in April to below 5%, cutting the headline CPI rate from 3.4% to 2.75% or lower (based on the one-third weighting of food in the basket and assuming non-food inflation remains at the current 1.7%).

Policy-makers are likely to wish to avoid a consumption-sapping rise in real interest rates, so a fall in inflation to well below the 3.5% one-year deposit rate could trigger an early downward adjustment in the spectrum of official rates.

Evidence to date suggests that the ECB’s subsidised loans to banks have been much less effective in delivering monetary stimulus than “conventional” QE as practised by the Bank of England.

A positive impact on the supply of money from such policies depends on the banking system increasing lending to, or buying securities from, non-banks (including governments). Banks “finance” a rise in their assets by crediting the deposit accounts of borrowers or sellers of securities with newly-created money*.

The ECB’s lending to banks via its weekly and longer-term refinancing operations rose by €512 billion between the end of November and the end of March. Bank lending to the non-bank private sector, however, fell by €70 billion over the four months, or €32 billion after seasonal adjustment. The decline, admittedly, might have been even larger without the ECB’s additional loans but it is reasonable to conclude that any positive impact was small.

The ECB’s actions have been much more successful in inducing banks to buy government bonds – purchases totalled €127 billion over December-March following sales of €39 billion in the prior four months. Comparing this buying with the €512 billion increase in ECB lending, however, suggests “pass-through” of only 25%.

Contrast this with the Bank of England’s QE2 operation, involving the central bank buying £125 billion of securities directly rather than relying on banks using officially-lent funds to do so. The monetary impact has been slightly reduced by small-scale bank sales of gilts – £15 billion between October, when QE2 started, and March. This still, however, suggests a monetary boost of £110 billion, implying pass-through of 88%.

The ECB, in other words, would have obtained a much bigger “bang for its buck” – three and half times the impact, based on the above figures – by channeling funds directly into government bonds rather than lending to banks. It could, alternatively, have delivered the same amount of stimulus with a much smaller increase in its balance sheet, thereby preserving ammunition likely to be needed in the event of “Grexit”.

April monetary statistics are released next week but the suspicion is that Eurozone money supply trends remain too weak to promote economic recovery, warranting further policy easing. The view here is that this should take the form of country-neutral QE, i.e. bond purchases spread across markets in proportion to GDP. Such an approach would deliver Eurozone-wide stimulus – helpful for “rebalancing” – while limiting risk to the ECB’s balance sheet and avoiding an inappropriate subsidy to peripheral sovereigns (implied by the operation to date of the “securities markets programme”).

*The “banking system” here refers to the central bank plus commercial banks. To the extent that newly-created money is held by overseas residents or governments, it will not show up in published monetary aggregates, which usually cover only domestic private sector holdings. Subsequent transactions, however, may result in the money being transferred to private residents.

Eurosystem gross lending to banks in euros rose by a further €7.1 billion last week to a record €1,339.5 billion, consistent with speculation that institutions in Greece and other peripheral countries are continuing to suffer an outflow of funding, necessitating increased borrowing from the ECB and national central banks.

The gross figure comprises “lending related to monetary policy operations” and “other claims”, under which the Greek and Irish “emergency liquidity assistance” (ELA) operations are recorded. The latter component rose by €4.1 billion last week, probably reflecting additional Greek ELA. This may have been related to last week’s announcement that some Greek banks had been temporarily shifted from regular lending (i.e. “related to monetary policy operations”) to ELA pending recapitalisation.

The switch, however, did not prevent a further €2.9 billion rise in regular loans last week, i.e. any reduction to the affected Greek banks was more than offset by an increase in lending to other institutions.

The suggestion is that that the ECB’s exposure to the periphery is continuing to mount, with a probable parallel rise in Bundesbank exposure to the ECB via TARGET2.

The fall in annual CPI inflation from 3.5% in March to 3.0% in April was slightly smaller than projected here – a 2.9% April print had been expected – and reflects a favourable “base effect” (i.e. an unusually large monthly CPI rise in April 2011 dropping from the 12-month calculation). Motor fuel price relief could maintain the headline rate at 3.0% in May, allowing a welcome suspension of the King / Osborne letter-writing charade. Shorter-term “core” price momentum, however, remains stubbornly strong and inconsistent with a return of inflation to target.

The lower headline annual rate conceals a solid 0.3% monthly rise in the core price measure tracked here – the CPI excluding unprocessed food and energy, adjusted for VAT effects and seasonal factors. The six-month rate of change of this measure (smoothed by taking a three-month moving average) remains near the top of its range in recent years, at 2.9% annualised – see chart.

A partial unwinding of the favourable April base effect in May could be offset by a recent reversal in motor fuel costs that promises to shave 0.2 percentage points from annual CPI inflation – the average price of unleaded has fallen back to about 138 pence per litre from 142.5 pence last month. The outcome may depend on rounding but this could allow the Governor to escape letter-writing duties – triggered by a headline print of 3.1% or higher.

The forecast here remains that CPI inflation will finish 2012 at about 2.75%, based on a small easing of core price momentum and broad stability of commodity prices and the exchange rate. Following the recent upward revision, the Bank’s central projection is similar (the May Inflation Report fan chart numbers will be published tomorrow).

Previous posts have discussed the US recession-warning properties of real narrow money (i.e. currency in circulation plus demand deposits divided by consumer prices). 10 of the 11 post-WW2 downturns identified by the National Bureau of Economic Research (NBER) were preceded by a contraction in real money. The exception was the 1953-54 recession, apparently caused by severe fiscal tightening as defence spending was slashed after the Korean war.

On the basis of this record, posts in 2010 and 2011 expressed scepticism about forecasts at the time – for example by John Hussman* – that the economy was about to enter another recession. Real narrow money grew robustly during 2010-11.

Real money has slowed since late 2011 but was still up by a solid 4.3% (not annualised) in the six months to April – see first chart. Recent posts, therefore, have suggested that US growth will fall back during 2012 while remaining respectable.

Monetary trends, however, bear close scrutiny. Real narrow money was flat between January and April. The latest nominal figure – for 7 May – was 1.8% below the April average, although the weekly series displays considerable volatility.

Also generating some concern here is a recent fall in the Monster index of online job vacancies – the decline in March and April was the largest since the end of the December 2007-June 2009 recession. The index turned down sharply just before the onset of that recession – second chart. (This weakness, however, is at odds with several other labour market leading indicators.)

At the time of writing, the Intrade prediction market was discounting a 16% probability of a recession in 2012, defined as two consecutive negative US GDP quarters between the fourth quarters of 2011 and 2012. The judgement here is that the probability remains well below 50% but is greater than 16%, especially given the risk of spill-over from an escalating Eurozone crisis.

*In his weekly market comment (link to left) of 28 June 2010 Mr Hussman wrote: “Based on evidence that has always and only been observed during or immediately prior to US recessions, the US economy appears headed into a second leg of an unusually challenging downturn.” The comment of 8 August 2011 stated similarly that “the composite of economic and financial evidence we presently observe has always and only been associated with ongoing or immediately impending recessions”. Mr Hussman now expects the NBER to judge that a recession started in May or June 2012, suggesting that one was not “immediately impending” in August 2011.