The Gradual Return to Fiscal Dominance

The UK government recently decided to transfer the coupon income from the Bank of England’s gilt buying program (Asset Purchase Facility (APF)) to the UK Treasury.  This is not only an interesting development in itself, but is yet another marker in the slowly evolving relationship between the central banks of the struggling developed market economies and their governments.
 

Looking at the UK’s recent action, there are two main effects. First, it will reduce UK’s debt/GDP ratio.  The transfer of £35bn initially and roughly £15bn each year after to the UK Treasury will have a material affect on the debt ratio, leaving it about 4% points lower in 2016 (all other things equal) than the IMF’s recent forecast (at 74%).

 

Source: Citigroup

 

Secondly, transferring the income has the same effect as buying more gilts as the Treasury has stated it will use the cash to enable it to issue less debt.  So, ultimately, the action results in there being less gilts in the market.  This is de facto QE, as Mervyn King, the BoE Governor, stated in his letter to the Chancellor acceding to the Treasury’s request.

 
Except, of course, this was not a request.  This is another incremental move towards ‘fiscal dominance’.  This is where the fiscal authority ultimately gains the upper hand from the central bank and we see monetisation of public sector deficits and debt.  This is the diametric opposite of ‘monetary dominance’, where the central bank forces the government to cut its primary deficit and stabilise the level of public debt outstanding.
 
It may be eminently sensible for the Treasury, in this case, to receive the interest income, especially as the Bank of England probably would have decided to reinvest the coupon income in more gilts anyway (as the Fed has done with MBS).  But what if it had not?  This would then have been the Treasury forcing the Bank to ease policy against its will.  A clear move towards fiscal dominance.
 
As Willem Buiter, Citibank’s Chief Economist, has stated, over history, fiscal dominance is the norm.  The previous 30 years gave birth to more independent central banks.  Governments were happy to allow this as these – as we now know – were the exceptional, NICE (non-inflationary continuous  expansion) decades.  Governments had little debt so there was no real harm – and indeed perceived benefits – in allowing independent central banks set interest rates to target (principally) inflation.
 
With the end of the NICE times and many developed governments running large fiscal deficits and alarming debt/GDP ratios, history is stirring, and there is a gradual return to fiscal dominance.  In Japan, the line is even less blurred as Ministry of Finance officials have recently attended Bank of Japan policy meetings, and government ministers have spoken openly about wanting easier monetary policy.
 

Large and persistent budget deficits, it has been well documented*, are associated with episodes of hyper inflation, as the following chart shows.

Source: Variant Perception

 
Furthermore, even periods of high, rather than hyper, inflation are associated with high government spending.

 

Source: Variant Perception

 

Already, most developed market countries are behind their projections for where they expect their budget balances to be, and almost none have made any serious progress in bringing their deficits down.

 

Source: Variant Perception

 

The policy of incrementalism towards fiscal dominance in the developed world has profound implications.  Governments will show less – not more – restraint with their budgets.  The prospect of higher inflation in the years to come is virtually assured, and the aberration of hyperinflation – which has the unsettling knack of taking root alarmingly fast – is no longer dismissible as the prognostications of fantasists.

 
 
* Monetary Regimes and Inflation, Peter Bernholz

Comments are closed.