This paper utilizes Critical Slowing Down (CSD; instability) indicators developed by statistical physics to analyse economic growth rate variability and secular stagnation in historical GDP data.
CSD measures instability, they say. And then, this:
The most commonly occurring pattern [in the GDP data] is characterised by an increase in CSD from the 1900s to 1940s, a decline in CSD between the 1930s and the 1970s, then a further increase in CSD from the 1960s to 2010. This pattern is significant in ~70% of the advanced economies.
An increase in CSD, a decline in CSD, and then further increase in CSD. Betcha that would match up well with an increase in finance, a decline in finance, and then further increase in finance.
Yes, and when I wrote the previous paragraph I was thinking of this graph from Thomas Philippon
 |
| Source: SSRN |
and this text from Philippon's FinEff.pdf:
The sum of all profits and wages paid to financial intermediaries represents the cost of financial intermediation. I measure this cost from 1870 to 2010, as a share of GDP, and find large historical variations. The cost of intermediation grows from 2% to 6% from 1870 to 1930. It shrinks to less than 4% in 1950, grows slowly to 5% in 1980, and then increases rapidly to almost 9% in 2010.
I want to put the CSD data and Philippon's together on a graph. But I
have to nudge it. The dates for the CSD are given in decades, while
Philippon's are given in years. And the CSD data is only
"increase" or "decline". It provides no
values that can be compared to Philippon's percentages. So the graph can only show increase, decline,
and turning points.
Here's what I came up with:
Don't
read too much into it. All I have to work with is dates. The upper and
lower turning points are just that -- turning points. The values are
not all the same for the uppers, nor for the lowers, as you can see on
Philippon's graph. And the two lines meet in 2010 only because that's
where the data stops.
However, the paths of the two lines are similar. Also, red always leads, and blue always lags. There might be something to this. Before finance (red) goes high, CSD (blue) shows stability. But as the size of finance increases, instability begins to increase. When finance is high, it peaks and begins to fall; but instability continues to rise for a time before it starts to fall. Finally, instability follows finance down (toward stability) and continues down while finance reaches its low point and begins to rise again.
This pattern makes so much sense to me that I think, given
more recent data, we'd see the blue line peak not with the red in 2010,
but perhaps a decade after.
I also find it interesting that
the red and blue are more widely separated near the lower turning point,
and less widely separated near the upper. The lag time varies,
depending on whether finance is low or high. The higher finance goes,
the sooner it impacts stability.
And yes, the lag time depends on finance, not on CSD, because finance leads and CSD lags. Finance is the cause, and instability is the effect.
Talk about not reading too much into a graph!
I added bullet points to this sentence from the article:
A wide range of ‘head-winds’ or inter-decadal drivers of secular stagnation are discussed by the literature, including
- public debt overhang,
- reduced aggregate demand,
- reduced innovation (supply),
- policy uncertainty,
- changes in demography,
- decline in education attainment growth,
- increased inequality,
- decline in labour productivity growth,
- decline in the quality of primary resources and
- changes in the structure and organisation of the financial sector.
I see reduced aggregate demand on the list, and increased inequality on the list, but I don't see the excessive size of finance that creates the inequality and reduces the aggregate demand. The structure and organization of finance is on the list, but not the size and cost of it.
Ignoring a problem doesn't make it go away. Just the opposite.
This part I thought was interesting:
CSD behaviour as outlined by Wiessenfeld (1985) and Wisel (1984) is best described with an abstract example of an oscillating system. Consider an object that is suspended between two springs, in a windy environment (the spring-object-spring system; Fig. 2a). The springs provide a deterministic restoring force (black arrows), and the wind provides a series of stochastic perturbations. As the wind pushes the object away from its equilibrium position the springs pull the object back, leading the system to oscillate about its equilibrium. If the springs are strong (strong restoring), the oscillations are fast and small. If the springs are weak (weak restoring), the oscillations are slow and large. If the relative strength of the springs becomes weaker over time, then the oscillations of the system transition from small and fast to large and slow (Fig. 2b).
This transition is characteristic of CSD behaviour.
Later, they come back to the "spring-object-spring" model:
The findings of this study suggest that CSD indicators may be useful for exploring inter-decadal macro dynamics. For example, the behaviour of the spring-object-spring system described in the introductory section is qualitatively similar to the behaviour of many non-equilibrium growth models. These include the Goodwin cycle, Minsky’s financial instability hypothesis and the Minksy-models researched by Keen (2013) and others.
Goodwin I don't know, but Minsky and Keen look at instability arising from finance. The size of finance, which didn't make the list of headwinds above, is surely relevant. And if you look at Figure 2b, the sine curve gaining amplitude over time shows the kind of growth that occurs in the financial sector of the US economy: relentless increase.
The problem is not that "the relative strength of the springs becomes
weaker over time". That's an analogy. The problem is that the relentless growth of finance gives rise to instability.
No comments:
Post a Comment