Yesterday’s autumn statement included the results of the Treasury’s study of the dynamic impacts of the cuts to Corporation Tax, which George Osborne is down from 28 per cent to 20 per cent. This study used the new HMRC Computable General Equilibrium model – as Fraser reported on Wednesday – and the results are impressive.
- The cuts will increase investment by 2.5-4.5% (£3.6-£6.2 billion in today’s prices).
- They will increase GDP by 0.6-0.8% (equivalent to £9.6-£12.2 billion).
- Given the share that we can expect to go to labour, that equates to an increase in wages of £405-£515 a household.
- As a result of higher profits, wages and consumption, we can expect the cost of the policy to fall by 45-60 per cent in the long run.
It is important to put the magnitude of those changes in wage levels in context. According to other, static analysis released yesterday, households in the middle quintile are £300 a year worse off as a result of the cumulative overall impact of changes in public service spending, taxes, tax credits and benefits since the Emergency Budget in June 2010.
That means the entire impact of the fiscal adjustment on family finances could – in the long run – be offset by the higher wages that result from lower Corporation Tax, although the Treasury will try to take a chunk out of those higher wages in tax as well.
There is an obvious political consequence to these findings: Ed Miliband’s plan to stop some of the planned Corporation Tax cuts would be a dreadful way to finance his fine objective of lower Business Rates.
And the findings in this report almost certainly understate the real gains from Corporation Tax cuts thanks to two critical limitations in the current version of the model.
It does not capture the effect of lower Corporation Tax in encouraging investments that embody new innovations in the capital stock, or any resulting technological spillovers. That means it does not capture the potential impact of Corporation Tax cuts on long run GDP growth, and only assesses the impact on the equilibrium level of GDP.
And the model itself does not capture the effect of lower Corporation Tax in encouraging foreign direct investment. Earlier research for the TaxPayers’ Alliance found that the effect on footloose international capital was likely to be one of the most significant results of changes to Corporation Tax rates. An ad hoc adjustment is applied to account for this but it would clearly be better if the CGE included at least a simple model of the international economy.
If the model included the effects of allowing the British economy to take full advantage of new technology, then it would almost certainly show an even stronger improvement in our economic fortunes with Corporation Tax cuts. Cuts in Corporation Tax probably increase revenue over time.
The report yesterday shows the urgent need to extend this welcome but faltering step from the Treasury towards proper dynamic analysis of policy decisions. Robust dynamic analysis needs to become the new norm for Treasury fiscal policy appraisal.
I think there are three steps needed to make that happen:
- Dynamic analysis using the HMRC CGE model should be required for all fiscal policy changes announced by the Government. It should also be used to generate key analysis like the graph of the distributional impacts of policy changes. Dynamic analysis should not be reserved for a single flagship economic policy.
- Each member of the Treasury Select Committee should be allowed to submit one fiscal policy for full assessment using the CGE model each year. That would allow for new ideas from outside government to access the same technical assessment available to ministers and officials inside government.
- The Treasury Select Committee should be empowered to commission regular (annual or biannual) reviews of the assumptions in the model and how it might be improved from independent economists. That would make it harder for the model to be adjusted just to fit the priorities of the government of the day.
Matthew Sinclair is Chief Executive of the TaxPayers’ Alliance.