Lies, Damned Lies, and Misleading Data

In one of last week’s videos (see below), I commented on the misleading UK media reaction to Q3 GDP data. I promised to expand on those comments here, and while I’m mindful of Thomas Carlyle’s description of economics as the “dismal science,” I hope you’ll bear with me. This is important—not least because how we measure the economy profoundly influences the decisions of policymakers, sometimes to the detriment of us all.

Let’s start with the Q3 GDP data itself. According to the ONS, the economy grew by just 0.1%—a sluggish number well below expectations and much slower than the 0.5% growth in Q2. Predictably, alarmist headlines followed, with words like “stagnation” cropping up in much of the commentary. But as I said in the video, this conclusion was wrong. The economy actually strengthened in Q3. That might sound odd, given what the data seems to show, but as Mark Twain famously said: “There are three kinds of lies: lies, damned lies, and statistics.”

The issue lies in how GDP is measured. The ONS primarily uses an output-based approach, which made sense back when the UK was a manufacturing superpower. It’s easy enough to measure the number of widgets rolling off a production line or the tons of wheat harvested from a field. But today, we’re a service-based economy, and measuring the output of an advertising agency, a bank, or a law firm is an entirely different challenge. How do you quantify the “output” of a fund management business? Frankly, I’m not sure you can (and I should know – we measured everything at WIM, right down to the number of cups of coffee made by our coffee machine).

A more reliable approach is measuring expenditure—how much households, businesses, and the government spend. This provides a much clearer picture of what’s really happening. And here’s the good news: on this basis, Q3 was strong. Household spending rose by 0.5%, government expenditure by 0.6%, and investment spending by 1.1%. Which begs the question, if all these components grew, why did GDP only rise by 0.1%? The answer lies in an oddity called “gross capital formation,” specifically a component labelled “acquisitions less disposals of valuables” (mostly non-monetary gold). This data series is notoriously volatile—so much so that the ONS describes it as “erratic”—and in Q3, it dragged down the overall number. Without this distortion, the data tells a far more positive story, with the three primary engines of the economy all growing.

Interestingly, UK labour market data released last week reinforces this picture. Hours worked in the economy increased by about 1% in Q3, yet GDP supposedly grew by just 0.1%. You don’t need to be an economist to spot that one of these figures is wrong. And it doesn’t take much guesswork to conclude that it’s the GDP figure. Measuring hours worked is straightforward; measuring output is, as we’ve seen, much trickier.

Of course, this discrepancy will inevitably lead some commentators to declare yet another fall in UK productivity. Hours worked are up, but output is flat—therefore, productivity must have fallen. But I would suggest that businesses don’t pay people more to produce less. This isn’t a sign of economic malaise; it’s a measurement problem.

This brings us to the broader debate about productivity. Between 1998 and 2007, UK productivity grew at an annual rate of 2.2%. Between 2010 and 2019, that fell to just 0.6%. The IT sector—a supposed driver of productivity—saw its growth rate decline from 14% per annum to 5.8% over the same periods. And in financial services, productivity went from growing at 4.8% per annum to contracting at 0.8% per annum. Is this credible? In some cases, over-regulation might be to blame—particularly in financial services (a concern which now appears to be shared by policymakers here and in the US). But across the economy as a whole, I strongly suspect we’re dealing with a measurement issue.

This matters because flawed data leads to flawed decision-making. Policymakers must focus on improving productivity, but they also need to confront the limitations of how we measure economic activity. Too often, these problems are swept under the carpet rather than addressed with fresh thinking.

Looking ahead, it’s likely that the erratic “valuables” series will mean-revert in Q4, resulting in a stronger GDP figure early next year. When that happens, I wonder what the media narrative will be. But for now, it’s worth remembering that the UK economy is performing better than the headlines—and, indeed, the official data—might have you believe.