Flying blind with ONS statistics

Back in 2022, during a deep dive into the Office for National Statistics (ONS) National Accounts data on Gross Value Added (GVA), I came across a strange mismatch in how retail performance was being reported. Before getting into that, it's worth explaining what GVA actually is.

GVA is a measure of the value a business or sector adds to the economy. For retailers, it’s the difference between the value of goods sold and the cost of buying those goods in. In practice, that means GVA acts as a rough proxy for profit margins in the industry. It strips out the turnover tied to wholesale, supplier and labour costs and leaves you with the economic output that's been created by the retailer itself – essentially what’s left after the cost of goods sold (COGS).

So here's where things got odd. The GVA data showed retail output (the purple line) was more than 10% below pre-pandemic levels in Q4 2019. But when you looked at the sales volumes data (dashed dark blue line), which track the actual quantity of goods sold (or the weight of shoppers’ baskets), output was down by less than 1%. That's a huge difference. If both data series are capturing the same sector, you'd expect them to broadly tally. Margins couldn’t realistically have collapsed by that much unless something was seriously wrong under the surface.

Eventually, the ONS revised its GVA estimates, and the gap closed somewhat. But even in more recent releases (the dotted purple line), the two measures haven’t fully converged. For me, this was the first sign that something wasn’t quite right in the quality or consistency of the data being published.

Those familiar with our Economic Monitor will know that a health-warning has been attached to Labour Force Survey (LFS) data for almost two years now, following falling response rates to the LFS. To be fair, this problem is not unique to the ONS. But a resolution on the figures keeps getting pushed out. First, early 2024, later on in 2024, and now we can expect them to come out in 2027. So that amounts to four years without official monthly labour market statistics – crucial for informing labour planning, and even more so in the setting of monetary policy.

The issues do not end there. Since March, we have been without Producer Price Index data, imperative in assessing the pass-through of global commodity price pressures into the domestic supply chain. The lack of these figures at such a crucial time of intensifying global shifts in trade patterns, and the associated commodity price impacts, leads to a real blind-spot in informing businesses of their upstream cost pressures.

Put simply, we can’t afford to be flying blind. With official data patchy, late, or under revision, decision-makers need alternative sources they can rely on in real-time. That’s why domestic, private and trade body-led measures have taken on far more importance than in the past.

The British Retail Consortium has attempted to plug the gap for our members, enhancing our Retail Sales Monitor (widely used and trusted across the industry) and developing more recently the HR Benchmark, a labour planning tool for retailers, utilising data covering one in three retail workers.

It’s no longer just about supplementing official figures. In many cases, it’s about substituting them. Until the ONS can consistently deliver accurate, timely and complete data, serious users will need to look elsewhere to make informed decisions.