Five to twelve on budget day. I’m stood in the chamber in my usual(ish) spot for PMQs. Back of the chamber, good view, nice post to lean back on. I glance at my phone, a flurry of messages pops up. The entire OBR document has leaked ahead of the budget. The Chancellor called the leak “deeply disappointing”. That was an understatement.
But this is neither the largest nor the most damaging mistake the OBR has made. The OBR has been making us poorer for 14 years, with a methodology that says public spending (cuts) don’t really affect growth all that much. That error has made it harder to invest in growth and pushed us toward a doom loop.
The OBR was founded by the Conservatives in opposition and then became a public body when they entered office. In 2010, the OBR’s forecasts initially showed that the deficit would close by 2015. By 2011, they said the deficit would close by 2017. Then in March 2016, they said it would close by 2018. Every time, the same excuse: growth was “weaker than expected“.
Conveniently, in the OBR’s model, the Conservatives’ swingeing public spending and investment cuts did not cause our low growth. It was always other stuff: the scars of the financial crisis, then Brexit, then Covid. But not austerity.
Welcome, my friends, to the doom loop. The OBR estimated that public spending cuts would not lead to significantly lower growth. So George Osborne made massive cuts to try and hit a borrowing target. But then public spending cuts did lead to significant falls in growth, leading to lower tax revenue and making it harder to hit that borrowing target. Then more cuts, lower growth, and lower revenues leading to more cuts, and on and on. No wonder we’ve had eight sets of fiscal rules in 14 years.
The doom loop is the main reason we had the lowest wage growth since the 1800s, stagnant living standards, a rising debt burden, low growth and flat productivity.
This doom loop was reinforced by the OBR’s central assumptions. Firstly, that current public expenditure – think teachers’ salaries, paying social security – has little impact on growth. In the OBR’s view, the impact of cutting £1 of current spending on growth (aka the “fiscal multiplier”), leads to only a 40-50p fall in economic output in year one and this effect falls to zero after 4 years. There is no long-term growth impact of public spending cuts in their view.
We know this is wrong. Cuts in health and social security spending made people sicker, leading to record high waiting lists and more people on sickness social security payments. But the OBR does not estimate these impacts. When I asked the OBR at the Treasury Select Committee about why this was, their response was that “it is by no means easy” to estimate this impact. It may not be easy to estimate, but that doesn’t mean the effect doesn’t exist.
You could also consider the two-child benefit cap. We know that poverty in childhood means lower grades. Kids who learn less today earn less as adults tomorrow. We know the long-term annual cost of child poverty is over £27bn per year in lower tax revenue and higher public spending. But when the Conservatives announced the two-child limit a decade ago, the OBR did not estimate these impacts.
When it comes to public capital investment (on stuff like roads and bridges), the OBR now believes this does make us somewhat more productive in the long run. Weirdly, they only came to this view in 2024.
But their view is that public capital investment has little impact on growth within five years. This is despite the real-world example of the United States. Biden’s large public investment led to a massive short-term economic boost after COVID, far outstripping other G7 nations. I asked the OBR about this too – surely this was a real-world example of large public investment leading to more significantly more growth? Their reply? “There is a bunch of reasons why the US might be different from the U.K.”
The issue is not that the OBR keeps getting its forecast wrong. In practice, trying to precisely forecast over £2.2 trillion of taxes and spending 5 years into the future within a £20 billion margin of error is impossible. It’s like trying to hit an apple with an arrow in a gale. The issue is that they keep getting their forecasts wrong in the same direction and not changing their judgement. If you keep missing the apple on the same side, you need to adjust your aim. If your model keeps pushing out the same error, something is wrong with the model.
I am not saying that public spending will magically all pay for itself through future higher growth. It won’t. But what I am saying is by not estimating the growth impacts of public spending, the OBR makes it harder for us to make responsible investments that will lead to future growth. Public spending, in the OBR model, adds a lot more to debt than to GDP. That led to irresponsible public spending decisions with too little investment.
The Chancellor made responsible choices to help us escape the doom loop Osborne created. She announced over £100bn of investment in clean power, new social homes, and rebuilding schools. She even managed to convince the OBR to score longer-term benefits of capital investment and of planning reforms. But note that “convincing” the OBR has no effect on the actual growth impact of public spending and policy changes. It’s weird they have to be convinced.
Leaking the budget was unacceptable. Getting the model wrong in the same way for 14 years is much worse. That mistake had led to lower growth and lower wages, making life far less affordable for us all. An OBR that properly counts the benefits of public spending (and the damage of cuts) would be a stronger, independent forecaster. It would help us make responsible investment decisions, making life more affordable for all.
[Further reading: Starmer stumbles over Reeves rumble]
Content from our partners
Related
