This research note is provided by Fathom Consulting. All of the charts below and many many more, covering a range of topics and countries on both the macroeconomy and financial markets are available in the Chartbook to Datastream users at www.datastream.com. Alternatively you can access Fathom’s Chartbook at www.fathom-consulting.com/TR.
Last month, Japanese sales tax increased from 5% to 8%, the first change since April 1997, when it rose from 3% to 5%. This week, we got our first insight into the impact of the tax change on both prices and quantities. It’s early days, but so far it is playing out pretty much as it did 17 years ago: inflationary in the short term; deflationary later on.
Around 70% of the Japanese CPI basket is subject to the sales tax. If last month’s increase were passed through immediately in full, CPI inflation would turn out two percentage points higher than otherwise. In the event, Japan’s core CPI rose from 1.6% to 3.2%, suggesting that the pass-through is largely complete. The fall in retail sales in April was substantial. In aggregate, retail sales fell by 13.7% (mom) – the consensus had been for a reduction of 11.7%. The drop in sales in April has been amplified by the fact that expenditure, particularly on ‘big-ticket’ items, had been brought forward ahead of the tax hike. The pattern of retail sales in the six months leading up to this year’s sales tax rise looks very similar to the pattern we saw 17 years ago. Sales drifted higher for several months before the hike, surged even higher in the month immediately before the hike, then dropped sharply in the month of the hike itself.
Following the 1997 sales tax rise, the Japanese economy entered a period of recession – output fell by 1.6% in 1998 after rising by 0.1% in 1997. How much of this was due to the sales tax rise, and how much to the fallout from the Asian Crisis remains an open question. We have used our own forecasting model, GESAM, in an attempt to assess the impact of an increase in the sales tax from 5% to 8%. We allow the model to run unhindered by any offsetting loosening of either monetary or fiscal policy. Relative to the baseline, where the sales tax had remained at 5%, we find that output falls steadily over a period of almost two years, and ends up just over 2% lower than it might have been otherwise. It takes more than four years for output to recover to the level that it would have been had the sales tax rise not been implemented.
Because we assume no offsetting policy changes – such as cuts in other forms of taxation, or an expansion of the QQE programme – the simulation set out in our chart is very much a worst-case scenario. And it should be stressed that it shows a projection for the level of GDP relative to a counterfactual world where the sales tax did not change. If we imagine that the economy might have expanded at close to its trend rate of growth (of around 0.5% a year) had the sales tax remained at 5%, then we find that it would take just over three years for the level of GDP to recover to its level before the sales tax rise. That is close to the amount of time that it took back in 1997. Evidence from our model, then, suggests that it was the sales tax rise rather than the Asian Crisis that did most of the damage back in 1997.
This year’s sales tax rise has undoubtedly come at a difficult time for the Japanese economy, not least because cutting interest rates to stimulate the economy is not an option. Nevertheless, if the Japanese economy does start to follow a path that is anything close to the one we have set out above, policy makers will respond, in which case we would expect to see a ramping-up of the QQE programme, some fiscal loosening (including cuts in taxation) and an increase in government expenditure – particularly on the investment side.
US – recovery on course despite disappointing Q1 GDP revision
The second estimate of US GDP for the first quarter, released this week, revised down growth by 1.1 percentage points to -1.0%. This was the first contraction in three years, prompting somewhat disappointed investors to push the ten-year bond yield well below 2.5%, an eleven-month low. Still, this data does nothing to alter our overall upbeat assessment of the US economy’s prospects for the rest of this year, supported by strong leading indicators and encouraging momentum in payrolls.
Consumption was the only component which made a positive contribution to growth in the first quarter. The dominant source of the downward revision was stocks: a dwindling rate of inventory accumulation subtracted 1.6 percentage points from GDP. At the same time, however, as companies operate with less of an overhang this should foreshadow a pick-up in production, thereby boding well for GDP growth in Q2.
At first glance, perhaps the most striking aspect of this GDP report lies on the income side, which showed a sharp decrease of some $213 billion in corporate profits to their lowest level in nine quarters. This amounts to only a blip, however, when one puts things into context: business profits as a share of GDP remain just 0.2 percentage points shy from the 11.1% record-high seen in Q4 2013.
In all, we continue to expect US growth in excess of 3% through 2014 as a whole. When the first, admittedly disappointing, estimate of GDP came out, we were “confident that the run of disappointing US data relating to activity around the turn of the year has largely been a consequence of the severe winter weather.” We still are. Leading indicators continue to suggest that the US economy has gone through a temporary set-back in the midst of a recovery that is otherwise more or less on track. The Markit PMI readings for May show real promise – services business activity had the highest reading since March 2012, while the manufacturing equivalent points to the strongest expansion in output for over three years.
Moreover, the trend in payrolls remains encouraging. The average of the monthly change in payrolls for the first four months of 2014 stands at 214k, 20K above the average for last year as a whole. As Fed policymakers noted at their last meeting, conditions in the labour market continued to mend and ‘participants generally expected further gradual improvement.’ Our statistical model, based on the weekly claims data, suggests that non-farm payrolls rose by 245K in May, 30K above the consensus. This rather strong bounce-back is driven by a reduction in the number of Americans that filed applications for unemployment benefits in May – the four-week average was at its lowest level since August 2007.
Mr Draghi’s turn to act unconventionally
Both Spanish and Italian inflation figures – the only ones out so far for the euro area – fell in the twelve months to May. The figures for Italy are particularly worrisome; whereas the consensus expected a slight increase in inflation from 0.5% to 0.6%, the actual figure came in at 0.4%. On a monthly basis, Italy fell into outright deflation – this month’s figure was the worst since the Euro was introduced in 1999. Clearly, these developments put further pressure on the ECB to act at its next meeting.
Fittingly, this week Mr Draghi re-iterated remarks relating to the circumstances that would lead the Governing Council to enact a variety of measures, ranging from further conventional easing to large-scale asset purchases. “An unwarranted tightening of monetary and financial conditions” would give reason for conventional monetary policy measures, i.e. a decrease in the policy rates by 10 to 15 basis points, which implies a negative deposit rate. However, as we argued last week , a 10 basis point rate cut is more or less fully priced-in by markets, implying only a weak – if any – impact on the exchange rate and government bond yields. Furthermore, according to our Global Economic and Strategic Allocation Model (GESAM), a reduction in rates of this magnitude would be far too small to have a meaningful effect on inflation and GDP.
If the ECB wants to counter “low-flation”, they will have to engage in unsterilised QE. According to Mr Draghi, “a broad-based asset purchase programme” would be triggered by “too prolonged a downward departure of inflation and/or inflation expectations” from the baseline scenario, thereby suggesting that the staff projections – to be presented at next week’s meeting – will be key. If the ECB were to revise their medium term forecasts significantly down, then QE would be a distinct possibility.
Mr Draghi also quoted ECB analysis suggesting that, in stressed countries, credit constraints (especially for SMEs) are already putting a break on the recovery, thereby adding to disinflationary pressures. This appears to bolster the case for the ECB to introduce additional measures – possibly including the purchase of ABS – in order to boost lending.
We consider purchases of government bonds most likely, with particular emphasis on Bunds and private debt instruments issued by residents of other euro area economies. As such, we see further downside risks to euro area sovereign yields, and Bunds in particular.