UK GDP grew by 0.3% in the second quarter, unrevised from last month’s preliminary estimate, but monthly output figures indicate a recovery in momentum through the quarter, suggesting stronger third-quarter expansion.

The 0.3% second-quarter rise reflected growth of 0.5% (0.54% before rounding) in services output, which accounts for 79% of GDP. Industrial and construction output weakened slightly last quarter.

The demand breakdown shows that GDP growth was driven by a rebound in inventories after first-quarter destocking along with a solid increase in public spending (both consumption and investment) and modest private consumption expansion. Business investment was flat both on the quarter and from a year earlier.

Monthly output data indicate that GDP was unchanged from its first-quarter level in April but rose by 0.2% and 0.4% respectively in May and June, reflecting services strength – see first chart. The June level was 0.35% above the second-quarter average, implying significant “carry-over” to the third quarter.


Slower GDP growth in the first half followed a fall in six-month real non-financial M1 expansion from a peak in June 2016 – second chart. Real narrow money growth bottomed in April and recovered in June, suggesting improving prospects for late 2017 / early 2018; July monetary data will be released on 30 August.

A post in December argued that official figures were understating the UK household saving ratio, probably reflecting an underrecording of income. It presented an alternative saving ratio measure based on data on capital / financial asset accumulation and borrowing – this averaged 9.0% over 2010-15 versus 8.0% for the official measure.

The Office for National Statistics this week gave a preview of revisions to be included in the national accounts annual update on 29 September. The new figures show the saving ratio averaging 9.3% over 2010-15 – higher even than the alternative measure calculated here (although this may also change significantly when the full new data set is released). The upward revision mainly reflects a correction of previous underrecording of self-employment income.

The chart shows the official saving ratio measure before and after the revision, together with the alternative measure. Revised numbers are not yet available beyond 2015 – a key issue is whether the sharp decline in the old official series in 2016-17 will be mirrored in the new data. The alternative measure is currently shown as rising in 2016.


A reasonable provisional conclusion, pending the 2016-17 revised data, is that household finances are stronger than previously indicated, implying a smaller risk of a significant consumer spending downturn.

The upward revision to self-employment income may also have implications for the MPC’s assessment of labour cost pressures. Monthly average earnings numbers cover employees only. To the extent that higher-earners have switched to self-employed status, extracting income through corporate structures, the average earnings series may understate growth of compensation per person across the employed population as a whole. It should be possible to estimate the size of any such effect when the new data set is released.

The view here that US economic growth will outpace consensus forecasts in late 2017 / early 2018 is based partly on the expectation of a significant contribution from the inventory cycle.

The Kitchin inventory cycle usually plays out over three to five years, with upswings lasting at least 18 months. Cycle lows are judged here to have occurred in 2009, 2012 and 2016.

Upswings follow a three-step pattern*, involving an initial burst of strength (A) followed by a pull-back (B) and final surge (C) into the cycle top.

The blue line in the chart shows the contribution of the change in inventories to two-quarter GDP growth. The contribution was negative for five quarters ending in the third quarter of 2016, consistent with a cycle low having been reached in 2016.


The growth contribution was positive in late 2016 but turned negative again in the first half of 2017. Steps A and B of the upswing, therefore, may have been completed, with the C boost scheduled to begin.

This interpretation is supported by an analysis of the inventories to sales ratio. The red line in the chart shows the two-quarter change in the ratio, plotted inverted. A rise in the ratio is usually followed by firms cutting the rate of increase or level of inventories, resulting in a negative growth contribution (and vice versa). The ratio fell sharply during the first half of 2017, suggesting that firms were surprised positively by final sales and supporting the expectation of significant restocking over the next several quarters.

*See Tony Plummer, Forecasting Financial Markets.

The MPC used the August Inflation Report to signal that it judges the market’s view of interest rate prospects to be too dovish. The market ignored the signal. One reason may be that the MPC’s stated view is being contradicted by a continued injection of liquidity via the term funding scheme.

Launched in August 2016, the scheme allows banks to borrow cheaply from the Bank of England for up to four years. The initial terms stated that the scheme would remain open at least until end-February 2018. The MPC this month decided not to extend it beyond this date.

Borrowing under the scheme has risen further from £69.3 billion at end-June to £80.4 billion, according to a weekly tally posted on the Bank’s website.

The Bank funds the scheme by creating new bank reserves. Total reserves, therefore, have continued to expand despite gilt and corporate bond purchases ending in March and April respectively – see chart.


A Sunday Times report claimed that “the cash has been used by banks to fund the surge in credit card lending and in unsecured personal loans”. This is misleading. For banks in aggregate, borrowing under the scheme has been matched by increased reserve holdings – the cash, in effect, has flowed back to the Bank of England.

By boosting liquidity and reducing banks’ need to compete for funds, however, the scheme has put downward pressure on deposit and lending rates. Lower lending rates, in turn, are likely to have stimulated credit demand.

The Bank’s critics, therefore, are justified in complaining that it is pursuing contradictory policies, i.e. warning of future rate rises and taking steps to discourage certain forms of lending while simultaneously providing additional “stimulus” via the term funding scheme.

A key mistake was to commit to an 18-month “drawdown window” when the scheme was launched – another example of the perils of “forward guidance”.

Global narrow money trends are giving a positive signal for economic prospects in early 2018.

Six-month growth of real narrow money growth in the G7 economies and seven large emerging economies (the “E7”), adjusted to exclude a contraction and subsequent rebound in Indian M1 due to “demonetisation”, is estimated to have stabilised in July after rising between February and June, based on data covering two-thirds of the aggregate – see first chart.


An earlier fall in real money growth between August 2016 and February is expected here to be reflected in a near-term loss of economic momentum, following a strong first half. The rebound in real money growth since February, however, suggests that the economy will reaccelerate from late 2017.

Growth of the G7 plus E7 aggregate was supported in July by a rebound in Chinese expansion. US and Japanese growth eased back but continue to give a positive message – second chart.


The final July reading will be influenced importantly by Euroland / UK data to be released next week.