© The Financial Times Ltd 2015 FT and 'Financial Times' are trademarks of The Financial Times Ltd.
The Tories must be cursing the Office for National Statistics. Just when they wanted to trumpet solid growth in the economy nine days before the general election, pesky official figures suggest growth of only 0.3 per cent in the first quarter. There is no doubt this is a bad figure. As the chart shows, it is the worst quarterly growth rate since the end of 2012.
The big action in the 2015 UK Budget comes in the moderation of public spending cuts. It doesn’t feel much like a rabbit out of a hat, but kills Labour’s charge that the ideological Tories are planning to cut public spending to its lowest level since the 1930s.
This is clear from the long run graph that total managed expenditure had been set to fall to extremely low levels (in green, the colour of the Autumn Statement). Though still true, spending now falls to a level just above the lows of 1999-00 and 1957-58.
Choices matter when measuring living standards. When looking at Britain’s living standards since before the recession, you can legitimately say incomes are back to the pre-crisis levels or still 10 per cent below those levels.
This post explains how (and the choices the Institute for Fiscal Studies made in its report this morning).
It showed UK living standards were back to the pre-crisis level. The IFS is an extremely reputable research organisation*. It also noted that the recovery in living standards has been painfully slow since the recession and household living standards still have not reached the 2009-10 peak.
I will show in this blog that definitions of living standards really matter, as do the choice of inflation measure, the choice of time period and the choice of average. It is perfectly possible and reasonable to arrive at a conclusion that living standards are 10 per cent below the pre-recession level with the same data as IFS used. This does not mean the IFS is wrong, but it has made choices that reduce the measured drop in living standards.
This is the IFS chart. Read more
On Wednesday, the Office for National Statistics for the first time published regional growth figures for the UK. The obvious question that popped into my mind was to compare the Scottish growth with that published by the Scottish government.
I produced the following chart and wrote what looked like a cracking story because the ONS measure showed little over half the growth of the Scottish Government data, raising questions over the strength of Scotland’s economy.
Scottish officials were naturally upset with the ONS calculation and these were “experimental” figures, so the Scottish government’s data is most likely to be accurate. The difference, however is so large that it requires a bit more analysis. The ONS put out a statement in response to the FT’s article saying it had gaffed. But the statement did not help much.
1. Forget the announcements
The net tax and spending measures are tiny compared with the forecasting changes to tax receipts. The purple bar on the right of this chart shows just how small the announcement are relative to the Office for Budget Responsibility’s forecasting changes. Read more
Market-sensitive data on economic growth, inflation and employment could be “drip fed” into the public domain because the Office for National Statistics’ website regularly failed to publish them at 9.30am.
Britain’s regulator of statistics has proposed that only the headline number for the most important economic statistics should initially be available online, with the full data sets published an hour later.
Industrial production and manufacturing figures for September were published by the Office for National Statistics this morning. They showed manufacturing picked up in September a little more than the statisticians had expected when they estimated third quarter economic growth, but not enough to trigger a revision to the 0.8 per cent initial growth estimate. The following five charts add some detail and put the figures in context.
1. Manufacturing still has a long way to recover
For all the talk of the “March of the makers“, manufacturing remains over 8 per cent below the economy’s 2008 peak and even further below the peak of manufacturing output which was reached in November 2000. Production, which also includes coal mining, oil and gas extraction, electricity, gas and water supply and sewerage has performed even worse since 2008, mostly due to the decline in North Sea oil and gas extraction.
George Osborne’s attempt to flatter the official headline measures of the public finances came unstuck on Wednesday after the UK Statistics Authority reprimanded the Office for National Statistics for treating the Treasury’s raid on the Bank of England as equivalent to tax revenues.
Upholding a complaint from the Financial Times about the ONS decision, the statistics watchdog called for a review of the headline measures of public borrowing and debt.
The decision will stop the chancellor from claiming borrowing is lower merely because funds have been moved from one part of the public sector to another, and demonstrates that the UKSA is willing to criticise the statistical work of its own government department as well as ministers.In November, Mr Osborne announced he would repatriate to the Treasury the funds building up under the quantitative easing scheme in the Bank of England arising from interest payments on the debt owned by the bank.
The Treasury hoped this raid would improve headline figures of government borrowing, helping Mr Osborne to meet his fiscal rules and say borrowing was falling, despite the weak economy.
Initially the ONS agreed with the Treasury’s arguments, but this has now been overturned by the UKSA, which said the counter arguments made by the FT were “more persuasive”.
With Britain’s economy bumping along the bottom and official data prone to large revisions, analysts have been on the lookout for indicators of regional and national recovery.
Until now, no one spotted a good set of figures, quietly collated at the Health and Safety Executive. Cranes puncturing the skyline are the most visible indicator of economic dynamism and have often been used to gauge the heat in economies as diverse as Dubai or China and the torpor of Tokyo.
The HSE started requiring the operators of tower cranes – used for new office buildings, infrastructure, larger residential blocks and big public sector projects – to register their addresses in 2010, providing an invaluable snapshot of the building industry in action.
The results show a remarkable degree of concentration in London and its surrounding counties, signifying a gulf between the optimism about a prosperous future for the capital and pessimism in the rest of the UK.
With London home to only one in eight people in the UK, it has seen more tower cranes notified to the HSE than all the rest of the UK put together. Almost eight in 10 cranes were in London, the southeast and the east of England.
When the Office for National Statistics relaunched its website last year, a geek like me was distraught as it had failed to make navigation of the UK’s generally wonderful statistics easier. “The new ONS website – aaaargh,” I commented.
Today, I gave evidence on the communication of statistics to the Public Administration Select Committee and to test whether the website had improved materially I thought I would pick a relevant question, to which I wanted to find an answer as a member of the public and as an expert user of the website. The question was: Is unemployment higher or lower now than in 1995?
This is the website journey (for lay and expert users) I described in the Committee, which I think is accurate and depressing. I don’t think a non-expert could find the interesting answer (at the bottom of the post) and an expert has to take a long journey to get there.
No one should be under any doubt. Jill Matheson, the national statistician, is consulting on changing the mathematical formula underpinning the venerable retail price index because the Office for National Statistics wants it changed. Consultations are not launched when experts think the status quo is fine.
At the heart of the issue is the realisation that the RPI formula is deficient and out-of-date. Continuing with the current method is the equivalent of Britain still thinking a 1970s Austin Allegro is cool, while the rest of the world is driving the latest Mini.
There is a chart regarding the UK economy which has become so ubiquitous it is known in our office simply as “the Niesr chart”, because it is often republished by the National Institute of Economic and Social Research. It is supposed to be a clear and concise account of Britain’s recent economic woes, putting the recession into accurate context of past recessions. It shows the current recession as the longest and nearly the deepest since the start of the 1930s. People don’t generally know that in the UK the 1920s recession was much worse, but I’ll leave that for now.
Here is the latest version of “the Niesr chart”, published today. Take a good look at it before I tell you why I have begun to become irritated by it.
It is arresting because it does most things right. It is simple to understand. It is clearly drawn and obviously in context. The problem is that that the Niesr chart might be showing us irrelevant nonsense. It is also not a sufficient description of the UK’s recession.
Britain’s traditional measure of inflation, the retail price index, is broken and needs fixing. The error goes back decades, has cost taxpayers billions, is still costing the exchequer about £1bn a year and has resulted in the rise in living standards being underestimated. Few errors in statistical compilation are quite this serious.
Find out how your living standard compares.
Select the range of years that contains your date of birth and watch as the graphic draws the spread of UK household incomes for people of your generation.
A few people have asked me for more data and information on the jinxed generation article in Saturday’s paper, which shows the youngest cohort of people entering the labour market were the first in over 50 years not to have higher living standards then their immediate forebears.
This post will be quite long and full of charts. It essentially shows our working. Read more