1. A Bit from the Past
While attending IIT Madras (July 1985–Jan 1987), apart from engineering etc., one of the subject areas that I tried to better get acquainted with was business and economics. So, first, I went through some books in the library at random, and in the process realized that before I could understand Finance and Stock Markets, I would have to study Management Finance. And, to understand that, first I would have to go, in the reverse sequence, learning: Management Accounting –> Business Accounting and Economics.
Since I also wanted to get a bit acquainted with the Indian situation (i.e. the Indian accounting etc. laws), I looked up, and settled on Grover’s accounting text as the main text. However, Indian texts were not very strong on conceptual explanations; for these, I came to rely on Meigs and Johnson, available with the IIT library. I found this book to be excellent. I even bought accounting journals and solved some problems.
Though I hardly remember any accounting theory today (and, somewhat surprisingly, am not at all interested in playing stock markets—here or in the USA), the one thing I still do remember is the fundamental equation: assets = liabilities + owners’ equity. Eventually, I also bought Jain’s Management Accounting and Prasanna Chandra’s book on Finance.
Actually, all this knowledge did come in handy very shortly later when I did a short summer course in the IIT on Entrepreneurship Development.
So far so good. The accounting and finance books, being concerned more with technical concretes, were fairly neat. But I found that the worth of the economics books ranged from “bad” through “very bad” to “loathsome”/”pathetic”. No exception. Stonier and Hague, I found, was somewhat bad. Paul Samuelson was from very bad to loathsome—but he was not pathetic.
Here, I remember a (good-natured) fight that I had with an IIT UG student, my hostel-mate, regarding the worth of Samuelson’s book.
My points: Samuelson failed to distinguish between a mixed economy and capitalism, and passed the first off as if it was the second. Further, his explanations sometimes fully inverted causality—making him a third-rate general thinker, granted all his technical brilliance (which I had assumed). Now, I had got it by reading Ayn Rand that in a proper economic system, there are only producers—no consumers. (I don’t recall here the excellent article in which she brilliantly traces the roots of inflation to irrational psycho-epistemology!)
His points: Samuelson was a celebrated author, from MIT, and a Nobel laureate. He was influential. These are the things that mattered. “Who reads your Ayn Rand at MIT?” was the sarcastic rhetoric with which he concluded his “position”.
With neither party willing to retreat, needless to add, we did not reach any conclusion in that debate—we just had to let it go at that.
Over a period of time, I have lost almost all my knowledge of accounting, finance and economics. I mean I am sure I could recover it comparatively very fast if I now begin studying. But I no longer find it so interesting any more. QM and computational science and engineering keeps me busy. … Which brings me to the next point today.
2. Computational Economics
Mid-1980s were the times when the PC had just begun making penetration. The twin revolutions of computers and communications, while they are still continuing, have by now become commonplace. I mean, even an undergraduate student these days invariably owns a powerful laptop—the way our “generation” had just begun owning those Casio hand-held calculators (the FX series).
Economists had always used computers, right from the days of main-frames. So, that way, computational economics is not a new branch.
But, now, there is a relatively new entrant in this field, called “Agents-Based Modeling” (ABM) for short. Personally, I find this kind of modeling both over-overdue, and highly interesting. [I also particularly love the fact that it reverses the acronym: “MBA” :)]
The idea in ABM is to perform a discrete kind of modeling that is focused on a finite number of economic agents and their interactions. Thus, instead of modeling some general economic principle in terms of, say, calculus-based(or algebraic equations, which in turn have been derived from some underlying theory, the new approach of ABM directly abstracts suitable economic agents, defines the rules of their micro-level interactions, and then, having represented all these in software, lets their interactions evolve in time during the course of a simulation. Thus, it is a bottom-to-top approach, which is good.
If you have seen or played with Conway’s Game of Life, or cellular automata in general, you will have some broad kind of idea as to the nature of simulation done in ABM.
Note, the bottom-to-top nature of such a modeling does not automatically make the process inductive, or the conclusions inductively valid.
But, yes, this kind of modeling, I believe, would be by nature better-suited if someone wants to inductively study economics.
The ABM field is not exactly nascent; the first papers seem to have begun appearing from late 1990s. However, the activity seems just about to take off, and for good reason.
I find it exciting (but only as a side-interest).
3. How Should It Be Done?
However, a simple Google search and a bit of browsing later, I realized how naive my excitement was.
Let me give just one example. Some ABM modeling done has consumers as economic agents.
Having read Ayn Rand, I find it offensive.
However, note that the field is still taking its baby steps. So, the current times are ideal if you wish to seize the opportunity and introduce a more rational kind of theorization in this field. Thus, for instance, the agents ought to be: producers, not consumers. And, then, perhaps, to make the model both realistic and representative of our times, you would have to add: “parasites,” “moochers,” “looters,” etc.
Which made me realize that I really don’t know how exactly such a thing should be done—i.e. modeled. Hence, this post.
If you are an economist (professional or amateur), here is a question for you to ponder:
Suppose you are going to write an Object-Oriented program (say in C++) for ABM. The question is: What kind of objects, their attributes, behaviours, relationships would you pick up (or isolate)? Also, what kind of rules of interaction (e.g. trade- or force-based)? why?
Just note, the idea is to try to make it a model of economics, not of accounting. It should be enough fundamental and abstract for it to qualify as a model of economic science.
4. One Observation
Just one more observation before we close. Suppose you have an abstract base-class (i.e. genus, in case you are not a programmer) such as, say, “Producer.” You may then derive (species) different sub-types such as “Farmers,” “UnorganizedLabor,” “FactoryWorker,” “Engineer,” “Designer,” “CEO,” etc. All these classes might implement a method (behaviour) named, say, “Produce().” The thing that is produced itself may have been derived from an abstract base like “Product” or even, in a certain sense, “Value.”
The interesting question is this. Whenever the stupids that are economists talk of people, they love to erect strawmen such as “Rational” man (and I have put the scare-quotes to emphasize that the word denotes their idea of an enforced sort of rationality) and “Rational” choices.
Since we are not stupid (LOL!), we will allow for Free-Will in our model. We might be able to incorporate such a thing in the attributes/behavior, and rules parts of our model. OK. For the sake of modeling, the degree of Free-Will may be derived either using some encoded rules—including using some random distributions (purely for convenience in implementation)—or even by letting real people make choices (as in software games that are played in teams).
The point or observation that I wanted to make in the passing is this: Any basic or fundamental economic modeling would have to consider volition explicitly—and ABM is great because representing volition can now be very direct.
This makes economics a very different kind of a science. Economics is a part of Politics, and studies social organization of men—individual men. Volition is necessary in economic modeling.
Volition (or any attribute or aspect of consciousness) is in principle outside of the scope of a physical science. On the other hand, it is in principle a necessary component of any discipline of humanities, including economics.
Bad philosophy leads to representation of men by calculus-based collectivized and deterministic rules in economic theory. Bad philosophy also allows a free reign to consciousness in fundamental physics (“observer” as necessary in physical “interpretation,” etc.)
5. Extended Scope:
Let me now add another side-point. If this ABM approach is sufficiently well developed, it could also show us how to model many other kind of interactions among men, e.g. those which fall in spiritual domain, i.e. with issues pertaining to consciousness, such as social bonding, religion, etc.
Think about it. ABM is not about just economics. It can be a powerful tool to model and observe in any kind of numerous or collective actions originating separately in individual men. Not only social structures but even spiritual interactions would be possible with it.
Before closing, here is the reminder that if you have suggestions concerning the questions raised in point 3. above, I would be interested in hearing from you—just drop me an email or a comment. Thanks in advance.
* * * * * * * * * * * * * * *
A Song I Like:
[A Note: BTW, note that when I pick a song for listing here, it is either mechanically at random, or randomly chosen by me. I never care for the cinematic or story or video or picturization context. In a way, I assume the radio-only or the record-player-/tape-recorder-only context of songs. Indeed, many of the songs I had never seen in any form until posting them here. Often, I find videos or picturization odd or funny—e.g. the “main ban-phool” song! In short, consider any video etc. context as omitted here.]
(Hindi) “phir wohi raat hai…”
Music: R. D. Burman
Singer: Kishore Kumar
[I know I still have to streamline my earlier posts… I will do it later this week or so… ]