US non-financial corporations' financial surplus – the difference between their retained earnings and capital spending – rose to 1.1% of GDP in the second quarter, according to flow of funds accounts data released yesterday. Excluding the third and fourth quarters of 2005, which were distorted by a one-off repatriation of foreign profits to take advantage of temporary tax incentives, the surplus was the highest since the fourth quarter of 1960.

The further improvement last quarter reflected a combination of stronger profits, cuts in dividends and fixed investment and faster destocking.

On top of this surplus, corporations raised cash from equity transactions for the first time since the second quarter of 2002, i.e. new issuance exceeded share buy-backs and retirements due to cash take-overs. Meanwhile, bond issuance remained heavy, though down from the record first-quarter pace.

Strong internal cash generation combined with capital market issuance allowed firms to increase their holdings of liquid assets and pay down short-term debt. The ratio of the two therefore rose sharply to its highest level since the end of 2006, supporting hopes of a strong recovery in capital spending – see first chart.

Some commentators have interpreted the recent contraction of bank lending to companies as supply-driven and likely to curtail business expansion. The comprehensive view of corporate finances provided by the flow of funds accounts suggests that bank debt repayment has been voluntary, reflecting the financial surplus and borrowing opportunities in credit markets.

The yield spread of non-investment-grade corporate bonds over Treasuries is inversely related, with a lag, to the sum of the financial surplus and net equity issuance, expressed as a percentage of GDP. The latter last quarter reached its highest level since the second quarter of 1958, suggesting scope for further spread compression – second chart.

Posts in early and mid April suggested that G7 industrial output would recover more strongly than forecast by the IMF and consensus. This was based partly on the "Zarnowitz rule" that "deep recessions are almost always followed by steep recoveries".

US business cycle history provides some guidance about the possible speed of a Zarnowitz-style recovery. There have been eight falls in US industrial output of 13% or more since 1890. In six of the eight cases, output regained its prior peak level within 21 months of the trough. The exceptions were the recoveries after the 1929-32 slump – when output plunged by 54% – and the 1944-46 recession, when the economy needed to restructure away from armaments production.

In the recent recession G7 industrial output fell by 19% over 13 months to a trough in March 2009. (The March trough had been signalled by monetary data in late 2008.) Based on the US evidence, discounting the post 1932 and 1946 recoveries, output could regain its peak level within 21 months, i.e. by December 2010. This would imply annualised growth of about 13% during the recovery.

The chart compares the actual rise in output since the March trough with this "Zarnowitz" forecast and an alternative 2% growth scenario. 2% is the maximum likely to be consistent with the IMF's latest published forecast of GDP expansion in advanced economies of only 0.6% in 2010. The last data point, for August, is an estimate based on US data released yesterday and a METI survey forecast for Japanese output.

Output is tracking slightly below the Zarnowitz path but is recovering much faster than forecast by the IMF and consensus. Based on the August estimate, the growth rate in the five months since the March trough has been 10% annualised.

Leading indicators suggest an acceleraton into the autumn, possibly closing the gap with the Zarnowitz forecast. For example, the OECD's G7 leading index, which is designed to predict industrial output, rose by 18% annualised in the three months to July.

Investors are debating whether the pick-up will be sustained into 2010. With the boost from the stocks cycle still at an early stage and financial market conditions improving, an optimistic view remains warranted but this requires confirmation over coming months from a stabilisation of labour markets and continued monetary expansion.

The three-month change in the stock of job vacancies returned to positive territory in August for the first time since March last year. Vacancies are a good coincident indicator and the 2% gain is historically consistent with quarterly GDP expansion of the order of 0.5% – see first chart.

The recovery in vacancies, as well as the prior plunge, was signalled in advance by the Markit job placements index. The Market index climbed further to a 17-month high in August, suggesting the vacancies rise will gather pace – second chart.

The MPC claims that a Bank rate of 0.5% is necessary to prevent inflation from undershooting the 2% target over the medium term. Its recent forecasting performance, however, casts doubt on its ability to predict near-term inflation movements, let alone developments two years or more ahead. If the Bank's inflation model is broken, it is fair to ask whether the pseudo-science of the Inflation Report fan charts should be ditched in favour of an ECB-style judgemental approach, including increased emphasis on monetary analysis.

In its February Inflation Report, the MPC predicted that annual CPI inflation would fall to 0.8% and 0.7% respectively in third and fourth quarters of 2009. A post at the time argued that this was too optimistic, with the Bank underestimating the inflationary impact of sterling's plunge during 2008.

Despite upside surprises in early 2009, the Bank actually revised down its third and fourth quarter modal projections in the May Inflation Report, to 0.7% and 0.4% respectively. Further disappointing outcomes, however, forced a significant change in the August Report, with the forecast raised to 1.3% for both quarters.

Two months of data later, this no longer looks credible. With today's 1.6% August result following 1.8% in July, inflation would have to plunge to 0.6% in September to average 1.3% in third quarter, as projected. The overshoot is likely to carry over to the fourth quarter. Bank of England Governor Mervyn King is probably now regretting his statement at the August Inflation Report press conference that a fall below 1% was "more likely than not" later in 2009.

A puzzle for the Bank's forecasters is that core inflation remains sticky despite a weakening of import price pressures as sterling has rebounded this year – manufactured import costs fell by 5% between March and July, following a 14% surge in the prior 12 months. The CPI excluding energy, food, tobacco and alcohol rose an annual 1.8% in August and would probably have climbed 2.4-2.5% in the absence of December's VAT cut (based on a National Statistics estimate that the reduction lowered headline inflation by about 0.5 percentage points). This would be the highest in its 12-year history – see chart.

As well as the size and persistence of the exchange rate effect, the lack of response of core trends to rising economic slack is troubling for the MPC's inflation optimism. This could reflect longer-than-expected lags but the Bank's forecasts may have placed overreliance on highly-uncertain estimates of the size of the "output gap" and its influence on pricing decisions.

The 1.6% August headline rate is in line with the inflation profile forecast presented in a previous post, although the breakdown is slightly different, with core prices higher and food prices lower than assumed. Updating inputs to take account of recent information, CPI inflation is projected to fall to 1.2% in September as a result of favourable base effects before rebounding to about 3% in January as VAT is raised back to 17.5%. The implied first-quarter average of 2.75% compares with a forecast of 2.1% in the August Inflation Report.

A post last week suggested that 4% UK broad money growth was sufficient to support a solid economic recovery because the velocity of circulation may be rising. Expressed differently, the demand to hold money may be declining as cash hoarding related to last year’s crisis reverses

One sign of declining money demand is a recent pick-up in mutual fund inflows. Net retail sales of unit trusts and OEICs totalled £14.0 billion in the first seven months of 2009, up from £3.9 billion in all of 2008, according to the Investment Management Association. Sales are on course to reach about £24 billion for the full year, well above the prior annual record of £17.7 billion in 2000.

If a £20 billion rise in mutual fund buying between 2008 and 2009 reflects a reduced demand for money, money supply numbers will understate the growth in cash available to finance economic recovery by £20 billion this year. This is equivalent to 1.3% of the MPC’s favoured broad money measure, M4 excluding cash holdings of “intermediate other financial corporations”.

As the earlier post noted, some monetarist economists argue that QE should be expanded further until broad money growth reaches 6% per annum or higher. With money demand likely to be rising by significantly less than 6% pa, however, this would risk creating excess liquidity, leading to new asset market bubbles and – further ahead – another pick-up in inflation.