“Data (In)dependent?”

The risks of ignoring UK private sector credit dynamics

The key chart

Trends in UK PS debt ratio (RHS) and PSC relative growth factor (LHS) (Source: BIS; CMMP)

The key message

Central bank decision-making that appears largely “data independent” in relation to private sector credit dynamics is unlikely to produce positive economic outcomes. Instead, it increases the likelihood of policy errors, especially in credit-driven economies.

It’s far too easy to blame certain individuals here (Bailey, Lagarde, Powell – you chose). The real blame lies more in the persistently flawed nature of macro thinking that frames policy decisions, however.

The fundamental error is to ignore that aggregate demand in a credit-driven economy (like the UK) is equal to income (GDP) PLUS THE CHANGE IN DEBT

(see Schumpeter 1934, Keen 2011).

The addition of the change in debt means that, in reality, aggregate demand is far more volatile than it would be if income alone was its source (as typically assumed). This is especially true for highly indebted economies.

In my analysis, I highlight three key factors when analysing the impact of debt on economic activity: the amount of debt, the rate of change of debt, and its rate of acceleration – each measured in relation to the level of GDP.

The chart above summaries these factors visually for the UK economy. The maroon line shows the UK private sector debt ratio (% GDP), while the blue line shows the rate of growth in credit compared to the rate of growth in nominal GDP (rolling 3Y CAGR).

The first key point is that the UK private sector has been deleveraging from most of the post-GFC period. The private sector debt ratio has fallen from 185% GDP in 1Q10 to 149% GDP in 2Q23.

The second key point is that a slowdown in the rate of growth can be enough to trigger a recession – an absolute fall in debt is not required. With the rate of credit growth below the rate of nominal GDP growth for the past three quarters, the impact of further policy tightening on aggregate demand is likely to be greater than forecast officially.

In short, the risk of policy errors continues to rise.

Data Independent?

The Bank of England (BoE) is likely to raise interest rates for the 15th consecutive time to 5.5% on Thursday 21 September. Policy makers will no doubt stress that inflation in the UK remains too high and that “raising interest rates is how the BoE can help to get inflation back down.” (see, “Why have interest rates in the UK gone up?” BoE website).

Note #1, that the BoE website highlights three variables that policy makers focus on when making decisions:

  1. How fast prices are rising now, and how that is likely to change
  2. How the UK economy is growing now, and how that is likely to change
  3. How many people are in work now, and how that is likely to change

Note #2, the lack of reference here to private sector debt dynamics. This is despite the fact that the BoE also argues that, “higher interest rates make it more expensive for people to borrow money and encourage people to save. Overall, that means people will tend to spend less. If people spend less on goods and services, the prices of those things tend to rise more slowly.”

Part of the problem here is that the impact of private sector debt on economic growth (variable 2) is largely absent for current macro thinking. Hence the assessment of how the UK economy is growing is likely to be based on traditional measures of income (GDP), while ignoring the impact of the change in private sector credit.

Why is this a problem?

The economist Schumpeter argued back in 1934 that, “in a growing economy, the increase in debt funds more economic activity than can be funded by the sale of existing goods and services alone.” Yet critically, traditional policy frameworks typically ignore the fact that:

Aggregate demand in a credit-driven economy is equal to income (GDP) plus the change in debt

As Professor Steve Keen argues,

“this makes aggregate demand far more volatile that it would be if income alone was its source, because while GDP changes relatively slowly, the change in debt can be sudden and extreme. In addition, if debt levels are already high relative to GDP, then the change in the level of debt can have a substantial impact in demand.”

Debunking Economics, Keen 2011

How CMMP analysis views UK debt dynamics

CMMP analysis incorporates a deep understanding of global debt dynamics. This includes not just the level of debt, but also its rate of change and its rate of acceleration – all measured with respect to the level of GDP.

Trends in UK PS debt ratio (RHS) and PSC relative growth factor (LHS) (Source: BIS; CMMP)

The key chart above illustrates these dynamics for the UK since 1983. The maroon line illustrates the private sector debt ratio – debt as a percentage of GDP. The blue line illustrates the 3Y CAGR in private sector credit versus the 3Y CAGR in nominal GDP (the CMMP “relative growth factor”).

The PS debt ratio increased from 68% GDP in 1Q83 to a peak of 185% GDP in 1Q10. The fastest rates of acceleration (peak excess credit growth) occurred in phase 1 (1Q83-1Q91) and phase 3 (1Q97 – 1Q01) as shown by the blue line above. In phase 4 (1Q01-1Q10), the debt ratio continued to rise but the rate of acceleration slowed.

The UK private sector has been deleveraging for most of the post-GFC period. The debt ratio fell to 149% GDP in 1Q23, the lowest level since 2Q02. My preferred RGF has been negative for the past three quarters. In other words, the growth in credit is slower than the growth in GDP.

They key point here is that an absolute fall in debt is not needed to cause problems, a slowdown in the rate of growth can be enough to trigger a recession (a topic that will be discussed in more detail in future posts).

With the rate of credit growth below the rate of nominal GDP growth for the past three quarters, the impact of further policy tightening on aggregate demand is likely to be greater than forecast officially.

In short, the risk of policy errors continues to rise.

Please note that the summary comments and charts above are abstracts from more detailed analysis that is available separately.