A popular saying attributed to Disraeli has it that there are three degrees of untruth: ‘lies, damned lies and statistics’. However, where statistics merely confirm prevailing policy assumptions, it is often politically expedient to avoid challenging orthodoxies, despite abundant economic evidence to the contrary.
Accurately assessing sub-national economic trends is often problematic, given the fragility of data and the leakage associated with intra- and inter-regional economic flows and supply chains. These problems are exacerbated in centralised states that have no effective regional political bodies to articulate policy concerns. They are compounded in the UK, where an asymmetric devolution settlement has given advantages, in terms of statistical development, to some regions at the expense of others.
The latest data from the Office of National Statistics (ONS) indicates that in 2017, in nominal terms, all regions of the UK except London and the south-east were less productive than the UK overall, with London being 33.4% more productive and the south-east assessed as being 7.4% more productive than UK levels in terms of gross value added (GVA) per hour worked. The fact that this data is nominal is the key factor here, since London and the south-east are the most expensive regions of the UK. The data is not, in fact indicating that the London and the south-east are outperforming the rest of the UK in terms of output as much that they are better at selling goods and services more expensively.
The principal locus of the financial services sector in London and the south-east also has a distorting impact. The output of the financial services sector is calculated using a method called Financial Intermediation Services Indirect Methodology (FISIM), which measures the GVA of financial intermediaries using the difference between interest rates charged and interest rates paid. The main problem with this is that it makes it seem that output has increased in risky scenarios, when the rate payable will increase – for example, banking output rose in the aftermath of the 2008 financial crisis as perceived risk rose. Some risk-adjusted estimates suggest that FISIM overestimates the value of financial intermediaries by as much as 40%.
The ONS has also recently started producing real GVA data that is adjusted for inflation, although this is only in terms of growth in the productivity series. This data shows a different picture, with London showing weaker productivity growth than other regions of the UK – although as it is only in terms of a growth index, it is not possible to see the base level this growth is coming from.
These differences in productivity can also be exacerbated by imbalances in government investment and spending, as can be seen below. Total spending per capita in London in 2016-17 was £943.66, more than twice the overall UK level of £435.31 and more than four times the level in the east Midlands, the region with the lowest spending per capita, at £220.34.
Simply bringing central government spending in line for all UK regions would do much to alleviate these imbalances in productivity.
The concept of an output gap, the gap between actual and perceived potential output, is a relatively simple concep. However its use, or more correctly its calculation, is often contentious, especially when applied to regional economies. Indeed, it is problematic calculating an output gap for a national economy, let alone a regional one, as there is a debate over how to calculate potential output. Without regional data on inflation, or even clearly defined geographic GDP, it is impossible to accurately calculate an output gap for a regional economy. Moreover, calculating an output gap is a political, as well as an economic, process as evidenced by the debate over the size of the output gap after the financial crisis and subsequent recession. Even now, estimates of the UK’s output gap range from -1.7% to 0.8% in January 2019. The International Monetary Fund suggests the following data is needed to calculate an output gap:
• capacity utilisations;
• labour shortages;
• average hours worked and hourly earnings;
• money and credit growth;
• inflation relative to expectations.
Moreover, there is a range of methodologies to estimate the gap which themselves producing a range of varying results. When attempting to calculate output gaps for sub-national or regional economies these difficulties are exacerbated by the paucity of data. This can be attempted to be offset by adopting one of the following approaches, including:
• calculate a potential output trend line utilising regional real GVA peaks to provide an estimate of the difference with actual output;
• adopt a production function approach;
• utilise Hodrick–Prescott filters, although is a heavily statistical rather than econometric approach;
• incorporate business surveys of producers to gauge capacity utilisation;
• Take a proxy, such as the unemployment rate as a measure of under-utilised resources, and then make a productivity assumption to get the output gain from employing all (or most) of them by reducing the u/e rate to a hypothetical maximum.
However, some regional bodies have attempted to calculate an ‘output gap’ using the UK’s GVA per capita as the ‘potential output’ level for the economy. There are flaws in this for several reasons. First, the regional data is included in the national data, so as regional GVA per capita rises, so will national GVA per capita. This also does not take into account the problems associated with calculating regional GVA, not least the amount of leakage between regional economies.
Without accurate comprehension of the structure and dynamic of regional economies, effective policy-making – both to offset structural weakness and stimulate real productivity growth – is problematic, to say the least.
West Midlands Economic Forum analysis, coupled with extensive anecdotal evidence, would suggest that in the case of the West Midlands, the regional economy is operating at around full capacity, given many of the transport bottlenecks present in the region, as well as the tight labour market conditions among skilled workers.
Bringing government investment across the UK in line with more subsidised regions such as London and Scotland would go some way towards alleviating the capacity constraints evidenced within the regional economy, with commensurate returns likely to be higher in the regions where there has been significant under-investment.
Accordingly, it is no longer (if it ever was) an academic question but a political one, and by determining public sector investment flows and private capital markets appetite, most notably FDI, it severely constrains regional economic potential. Accurate statistics in themselves will not immediately reverse such flows, but at least they can start to shift deeply ingrained, if stereotypical, perceptions of some of the potentially most dynamic regional economies globally.
Rebecca Jones is an economist and Paul Forrest head of research at the West Midlands Economic Forum, an independent forum bringing together representatives of the public, private and voluntary sectors to evaluate real trends in the local economy.