Applied Feedback systems–Economics

Sorry it’s been a few days.  I had my 20th year High School Reunion to attend to, and that took up most of my energy for most of the last week…

In any case–this post was inspired by a thought that my brother M. sent me.  M. would probably label himself a libertarian or a moderate-to-liberal Republican.  In any case, M. works in finance in a manufacturing firm in Chicago and gets to see first hand both how private enterprise works, but also how laws and taxes play a huge role in shaping how private enterprise does what it does.

What he sent to me was an extended thought about how American business–especially the leadership in the form of CEO’s and Board of Directors–has really gotten off track recently.  Although he doesn’t doubt that CEO’s deserve a lot of money for the hard work that they do, he also thought that the incentive structure and the checks and balances (like a Board of Directors that was in a truly adversarial rather than colluding stance to the CEO) were totally out of whack in many American enterprises and that this needed to be fixed.  Overall, he reflected that he sounded a lot more like his youth, when he tended to have a very idealistic sympathy for more socialist ideals.

In any case, it got me to thinking about the economy and how gov’t and private enterprises structure this realm.  My direct response to him was in a message that stated the following:

What I would say is that idealism at either end of economic spectrum is misguided. Theory takes over too much and starts to lose its grip on the reality of how people actually practice business and economics on a concrete level… and that’s what has gone wrong.

Gov’t control of everything–or corporate CEO control of everything–both of these situations lead to very few people attempting to control a system of 100’s of millions of people doing whatever they want–and that’s just hogwash.

The best systems will have high user participation–not only in low level activity–but also in higher end control and leadership. By doing this, you have a greater chance to have more good ideas take over–or at least have more of the stupid ideas get shot down before they do too much damage. This is what regulated market capitalism is. Gov’t acting as a balancing weight between different business groups (small bus. large bus. labor, etc..) and also doing work to make sure the foundations of the market (infrastructure, law, security, etc) are strong enough for all the raucous (and productive!) activity going on within it.

In essence–it’s like democracy in politics. Democracy–for all it’s ugliness at times–is better than all of the alternatives because it does give a greater percentage of the population a chance to put their say in… and the less democracy we have–the more likely we are going to have idiots take over and run us into the ground before we can all apply collective (if belated) brakes to the situation.

Now.. what does this have to do with feedback systems??? Well, the real complexity of the system and some of the methods of control (both systemic/passive and agent-oriented/active) are intimately tied up with different kinds of feedback systems.  In addition, the fact that the overall economic system is a very complex combination of these feedback systems is part of the reason that underlies my claim of why you need the broadest group of conscious actors to be part of the “coordinating” group of people involved in economic activity.

Some examples and evidence to explicate what I’m talking about:

1) Overall, the way that most people understand a free market is through the basic process of supply and demand, which determines production amounts and prices.  At their root, these systems are negative feedback systems, in that they tend towards self-correction and hone in on stable points given a certain set of external factors and circumstances.  Concrete point–if there is a bumper crop of wheat in a year, then the supply increases, the price drops so that the “excess” production will be purchased, and the system regulates itself eventually.  If the next year, the production is especially low, then the price will rise to compensate.  Such is the life of a free market with clear price signals and where demand and supply are relatively flexible.

2. Although the situation above is seen as the cornerstone of all economic transactions in the free world, it is not necessarily true that this is how all economic transactions are achieved.  Positive feedback situations exist throughout the economic system and are actually quite essential to a lot of the profit-making that is done there.  Interestingly, the majority of these kinds of situations are tied to information processing.  While strongly-material-based economic transactions are closely related to negative feedback systems, information-based transactions are often more of a positive-feedback type character.  To be concrete–think of the initial offering of Netscape’s stock.  As the price rose, more people wanted it.. this is the reverse of a negative feedback system, where, if the price rose, less people will purchase it.  Instead, people view the price increase as a sign that the product is of greater worth, and they think they can sell it for even more money, so they are willing to pay the price.

Such kinds of economic activity are rightly called “bubbles,” because they do act like physical soap bubbles in that you can blow them up to ever greater sizes until the physical reality of gravity and structural strengths of soap have been been pushed beyond their limits and the whole structure collapses.  Then, the $600,000, 2 bedroom house that you purchased because you thought it would appreciate 30% in value in the next year turns out to be worth ony $150,000, and you are screwed.

Such is the life of positive feedback systems. It can be great if your timing is exquisite (or if you have inside information or are in control of government regulation…), but if you are not in on the joke or are just unlucky, then you will easily get fucked.

3. There are other positive-feedback instances that are not quite so evil.  For example, the QWERTY keyboard layout (although slightly evil in that it produces lower typing speeds on purpose, see here…) became the dominant keyboard layout through positive feedback systems because it became relatively more dominant early on, which made it relatively more useful for everyone to choose it over other typing alternatives. Thus it was self-reinforcing to learn it, because as more people learned it, it became ever more attractive and useful to learn it and became ever more the de facto standard.  This quickly drove its competitors into extinction…    This kind of positive-feedback process is not necessarily a bad thing–it leads to quick standardization, which can improve efficiency–but it also does not guarantee optimal results as a negative feedback system (as we saww in the types of cases of increased grain output and the like…).

4. Positive feedback systems are also important in generating enthusiasm for early technological offerings.  Often a well-worn technological system that has all the kinks worked out will be dominant within a society–as it should be–but there might be another system that is technically more expensive/less efficient at first even though it shows great potential to be be superior in the long run… For example–DVD’s were 10x as expensive and VCR’s at first, without really a 10x improvement in quality, but enthusiasm for the product led to investment which lowered the cost of DVD players dramatically to the point that you can hardly even find VCR’s anymore.  This entire process was based on human enthusiasm for novelty and a belief in the superiority (and the ability to lower the price) of the new technology.  If it had, for example, been impossible to lower the price of DVD players such that they still would have cost over $1000, then it would not have succeeded, despite the fact that all those earlier purchasers thought it would.  Rational decision making–based on negative feedback systems–would not have predicted this kind of outcome, because it doesn’t deal with enthusiasm in productive ways.

Okay–those are the examples of feedback systems in the economic system that I think are directly relevant here.  As one can reflect upon, they are really quite ubiquitous throughout the economic system and play crucial roles in most of the economic activity that happens in our society (and all others…).

Now the main argument that I referred to at the beginning deals with these things in the sense that–as should be clear now–taken together it is entirely unreasonable to believe that a small group of people–no matter what their resources–are capable of grokking all of the complexities of the interactions of these positive and negative feedback systems in any way that is going to be optimal over the long run.

We are not gods.  Not even with computers.  We are merely mortals and there is no pre-configured story that you can know in advance.  Stuff is going to happen–disturbances exist that will set off positive-feedback loops that you have never contemplated–and the whole projection that you thought would guarantee the best outcome–it will be fucked.

So don’t try to be god.   Just be human.  Do your best and enlist all the best people you can and as many as possible and work through all of this.  Build systems that push people to be involved, to use their brains, and to strive, and you will be more likely than not to succeed better than any system that excludes or ignores such constraints.

And that’s all I have to say right now on the matter…


About Prof. Woland

I contain multitudes. Come meet us.
This entry was posted in Uncategorized and tagged , , , , . Bookmark the permalink.

One Response to Applied Feedback systems–Economics

  1. Pingback: Information, Filters, and Power | The Philosophy of NOM

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s