FUTURE VERSION OF JAVA SIMULATION

The Java simulation of Lonergan's macroeconomic dynamics has two aims:

(1) to provide a visualization that may enhance the compensating features of a diagram "that a system of simultaneous equations may imply but does not manifest" (p. 53). [N.B. It is not clear whether any specific "system of simultaneous equations" is implied in, or can be derived from, the series of equations from equation (2) to equation (47) listed on pages 48 to 160.]

(2) as a tool in the form of a computer game with which players can test their understanding of Lonergan's insights into dynamic equilibrium .

A future version will attempt to integrate the Diagram of Flows with the Pure Cycle of the productive process in Fig. 27-1 on page 150..

The Diagram of Flows will display the circulation of payments with 15 sliders whose values will be set by each of the 15 players according to a set of rules to be constructed.

The sliders are: I', I", O', O", RF, E", E', S", S', D", D', c", c', i", i'.

These 15 parameters are named in the diagram. Three of these yield w (the ratio of surplus income (I") to total income (I’+I") per interval). This will be the basis for deciding what fraction v (from 0 to 1) of the surplus expenditure E" will be spent on "new fixed investment" (p. 146) at a given "interval."

An interval can be a year or a quarter or a month, etc., in accordance with accountants' practice. A series of intervals constitutes a "phase" with "certain defined characteristics" (pp. 113-15) that have been empirically verified (e.g. the acceleration of surplus production is greater than that of basic production, expressed symbolically as |dQ"/Q"| > |dQ'/Q'|).

How and when should the decision to decrease the fraction v be made to maintain a pure cycle in the productive process?

According to Fig. 27-1 on page 150, it should be made sometime during phase 3. The fractions v and w keep changing with time and their graph in rectangular coordinates have both magnitudes and directions; the directions can be changing upwards or downwards, smoothly or sharply; there can be an angle between the two directions and its cosine can therefore influence the magnitude of f.

The magnitude of f can have an "ideal" maximum, a "slightly premature" maximum or an "extremely premature" maximum (pp. 149-50). The "ideal" f-max connotes only expansion and no contraction. If the movement is from basic expansion to basic contraction (indicated by a decreasing index [or weighted average] of basic quantity Q'), it connotes a "slightly premature" f-max. If the movement is from surplus expansion to surplus contraction (indicated by decreasing index [or weighted average] of surplus quantity Q"), it connotes an "extremely premature" f-max due to great "overexpansion" or deceleration of surplus producton Q". (It is quite difficult or perhaps impossible to visualize these movements with NON-ANIMATED graphics.)

The empirical verification of phase characteristics can only be made through statistical analysis of global databases in banks, firms, households, etc. Such analysis may take many decades to reach some reliability. [On page 55 of Method: A Journal of Lonergan Studies (15, 1997), Burley expresses a "need to think in terms of a century of research and education.]

In the meantime, it may be heuristically fruitful to posit hypothetical situations and study their consequences with animated Java simulation. How? (We would appreciate suggestions from our visitors if any.)

--------------

VALUABLE SUGGESTION FROM EILEEN DE NEEVE, Oct. 1, 2003

The applet of Lonergan's diagram of monetary flows is an excellent idea and
I wonder whether I might make a suggestion. Would it be possible to define
the relationships among the variables first in an economy that is in balance
but not growing (a static phase) and then in a pure cycle, which is
Lonergan's notion of well-functioning economic dynamics. The pure cycle
would allow for three steps: a proportional expansion, a surplus expansion,
a basic expansion. If the three-phase cycle is too difficult, the
proportional phase could be omitted. This simulation would establish the
existence of a pure cycle in which there are changes in production but no
speculative financial booms or recessions (no money shocks).

In a pure cycle your equations would be as follows:
Basic demand (I') = c'O + c"O" (D' = D" = 0 in a pure cycle)
Basic supply (O') = E' + S' (In a static phase S' = 0)
Surplus demand (I") = i"O" + i'O'
Surplus supply (O") = E" + S" (In a static phase S" = 0)
The ratio of the size of the two sectors is more or less O" : O' = 1 : 4.

Redistributive function (F) = S" + S'
The redistributive function is the source of CHANGES in the money flows in
the economy. In a static phase there is no change in the money flows
through supply and demand (I", I' and O',O'); i.e. no net flows to or from
the redistributive function (S' = S" = 0). However, money flows must
increase in the expansion. This occurs in positive net flows through S' and
S". Demand will then increase through increases in income as a result of
expanded production. Perhaps the applet could show the changes in the pure
cycle? Would S' and S" be choice variables?

What starts the expansion could be the beginnings of new ideas (Insights)
which require financing from R (the source of bank, government or
international loans or investment). So S' and S" flows from R to O' and O"
increase to allow a proportionate expansion in both basic and surplus
circuits. Assuming the parameters c and i do not change (about 0.78 and
0.22 respectively) the system expands. According to Lonergan, in a
proportional expansion the growth rates of the basic and surplus sectors are
the same.

In a surplus expansion, the basic (consumer) sector can only expand with a
lag because it needs new capital goods and services to expand its
production. So the flow from R to O" (i.e. S") increases, making O" growth
rate > O' growth rate. However, this affects c"O" and the consumer price
index rises while the output of consumer goods O' does not. In your diagram
which is in money units, this would show an increase in O' (The ratio O'(t)/
O'(t-1) would indicate the price change.) The rise in the consumer price
index creates profits which increase the flow i'O' to match the increased
flow from c"O". A possible difficulty is that in this phase c'O' should not
change so that the balance of O' flows to i'O'. Maybe this could be
achieved by a fall in the c' and a corresponding rise in i'?

In a basic expansion, other things being equal, the rise in consumer prices
would bring producers into business to produce consumer goods. The growth
rate of O' would rise while the growth rate of O" would move towards zero
because the new capital projects have been completed and have spread through
the economy once the construction lag is over. The cycle ends with a return
to the static phase as the growth rate of consumer goods reaches a maximum
and falls to zero. Alternately the output of consumer goods O' is as great
as it can be given the state of resources needed for production.

Perhaps the applet can handle this by having a maximum value for the money
flow (S") from R to O" with the flow shifting to O', with the smoothly
increasing and decreasing values of S" and S'reflecting the lags. I guess
this would require two or three period lags to allow for the rise and fall
of net flows from the redistributive function, in each sector? The values
of S" and S' could be about 0.025, 0.05, 0.025 of the values of O" or O'
over each three period phase.

I realize that you now have many new commitments and I am not certain that
these changes can be added to the applet, but it would be important for
Lonergan scholars and for economists in general to be able to visualize
Lonergan's notion of a pure cycle as a macrodynamic ideal. How the ideal is
not achieved in reality can then be explained in various ways. I hope you
or one of your colleagues might find it possible to pursue this. If I can
help in any way I would be glad to do so.

With good wishes to both Father Marasigan and yourself,

Eileen de Neeve
Thomas More Institute for Research
c/o 3405 Atwater Ave.
Montreal, QC, H3H 1Y2, Canada

(N.B. The suggestion above was made by Dr. Eileen de Neeve to Dr. Luis Sarmenta about the Java Simulation team-constructed before the 1999 publication of MD, and now in need of many modifications. VM. Nov. 5, 2005)

--------------------

Community-centered | Indeterminacy |

| Emergent Probability & MD |