On Mon, 29 Apr 1996, Steve Keen wrote:
> Duncan comments on my brief description of a multi-sector model:
>
> >As you describe it, I don't see how your model addresses the problem of
> >the valuation of the bank liabilities. Did you implicitly assume a
> >constant price level? In fact, your model sounds as though it is an
> >example of a credit system operating with a given value of bank
> >liabilities in terms of commodities, where the bank reserves are
> >absorbing the violations of budget constraints in disequilibrium.
>
> No, the price level is a function of wage and commodity input costs and a
> competitively set markup:
>
> P[t+1] = (w[t]*e[t]+p[t]*A)(I+M[t+1])
I assume that the money wage adjusts in response to the tightness of the
labor market. This is the "second approach" I referred to in my earlier
postings, where one regards the value of the government debt as in a
neutral equilibrium, so that it is subject to purely historical forces. I
think there are some problems with this, concerning modeling the
bargaining process by which wages are actually determined, but it
certainly is a possible way to put together a consistent model.
What this model doesn't address is why agents are holding this bank money
at all. One can give a good answer to this question in real-life
historical terms by referring to the credit of the state, but the
equations you write down, while they determine a path for the price
level, don't consider alternative institutional structures. This isn't a
purely theoretical issue: in many parts of Latin America and Eastern
Europe the dollar circulates as a de facto money competing with the
currency issued by the nation states, and many bargains, including some
wage bargains, are struck in dollar terms.
Duncan
>
> where the markup is a sigmoid function between a minimum (above the interest
> rate) and maximum and reflecting sectoral profit rates.
>
> The point is that there is no presumed link between the value of bank
> liabilities (which is the product of an accumulative process over time) and
> the prices of commodities (which is the product of a non-accumulative
> process over time). Inflation can occur (because of wage claims, mark-up
> adjustments and inter-sectoral price effects), thus devaluing bank balances
> in current terms. That said, I haven't been able to simulate the model yet
> (have to write some software). But at the conceptual level, there is no link
> between current commodity prices and current bank balances (except in the
> ex-post sense that current prices can be used to deflate current bank
> balances [if an index is developed, which is an issue in itself!]).
>
> Cheers,
> Steve Keen
> Steve Keen
> Senior Lecturer
> Economics & Finance
> University of Western Sydney
> PO Box 555 Campbelltown NSW 2560
> Australia
> s.keen@uws.edu.au (046) 20-3016 Fax (046) 26-6683
> ________________________________________________________________
>
>
>
>