Perhaps the key point differentiating the PSST view from a more traditional AD-AS view is assumptions about the speed at which new market connection are made and factors relating to the co-variance in that speed.
If connects were instantaneous then we could just retreat to Walrasian Equilibrium and be done.
If connects are not instantaneous then first we have curious question about co-variance. Why are so many connects broken or made at the same time. Further, why is it that sometimes it seems rapid change is making connections faster than breaking it and other times its going the other way.
A simple appeal to well, sometime things are like this, sometime they are like that, will not do. The network is massive. Connections happens every day. When we look at employment data we are looking at the mean number of connections over some period of time. The central limit theorem should apply.
That is overwhelmingly things should just move along as if we were in a smoothly rotating economy. Not because we are, but simply because the number of changes is so massive and so spread out randomly.
We have to appeal to major systemic shocks in one form or another to get around this. Or, perhaps some sort of collective reallocation of attention, though I am not sure how far you can go with that.
Yet, this brings up the role of policy and liquidity. Suppose that credit “greases the wheels” of connection generation. Its not hard to imagine how. Capital is easier to finance, start-up costs easier to obtain, bumps in the road easier to smooth out.
Now are we moving to a synthesis where increased connection generation is the transmission mechanism for monetary policy?
6 comments
Comments feed for this article
Monday ~ June 13th, 2011 at 3:09 pm
Lord
In a balanced, diversified economy, it would take a systemic change to affect it, but there can also be unbalanced, undiversified, economies where the largest share of profits come from only a few large industries and whose shifts can tilt the whole economy. In the last cycle it would have been housing, debt, and finance that became outsize for the economy, a one trick pony. The question then becomes whether each of those other industries growing only little faster can’t make up for it, or they are all so marginal that only another pony can make up for it.
Monday ~ June 13th, 2011 at 4:18 pm
Corporate_Serf
Phase transitions? Not an economist, so perhaps missing the point of it entirely, but in random graphs, for example, you connectivity/spectrum of the graph can have sharp [hase transitions
Tuesday ~ June 14th, 2011 at 12:47 am
Greg Ransom
Before you know it, you will have reinvented Hayek.
Exending the time length of production processes & then clipping it short is just the break you need.
Consider the long term production good _housing_ …..
Now many hints do you need?
Tuesday ~ June 14th, 2011 at 11:14 am
Hyena
I’ve made your point about credit–it improve the speed of trial and so discovery–several times.
Tuesday ~ June 14th, 2011 at 3:51 pm
Wonks Anonymous
Bill Woolsey on PSST. I don’t expect it will cause Kling to reconsider, since he never seem to give any reasons why a PSST story is more plausible than any other or why AD+AS is less plausible, he just says “I have lost the macro religion” or “I want to promote this view” and leaves it at that.
Tuesday ~ June 21st, 2011 at 1:08 am
Patrick
Sure, there must be shocks. Quite trivially, the economy was shocked. But we needn’t appeal to exogenous shocks, as they’re often treated in modeling. And the shocks need not be systemic either. Again, they’re trivially “systemic” in that the end up having consequences for the whole system, but they needn’t be systemic in the sense that they’re transmitted to a large portion of the system simultaneously(or nearly) through some coordinating mechanism. (ie liquidity, interest rate).
Systems of differential equations in physics and chemistry often show emergent behavior: Tipping points. A realignment starts gradually, and locally. It isn’t much of a stretch to imagine it accelerating. As it spreads to each person’s local network of connections, they decide not to take the risk of making new connections, based solely on local conditions. That decision makes local conditions worse for other people in their networks, making other people in the network less likely to to form new connections. Thus causes the shock to spread as well as feeding it back to the originator.
The chemical reaction in this video stays stable for a time, then undergoes a rapid transition. It doesn’t require any exogenous shock and the cause is not anything systemic. It is a cascade emerging purely from localized interactions.
Human behavior is at least as complex.