3 ms·

First, I really appreciate author trying to understand economy better in this way, even though this simulation is quite naive. I think everybody should do that.

by js8 3y ago

First, I really appreciate author trying to understand economy better in this way, even though this simulation is quite naive. I think everybody should do that. But as others have noted, it has no concept of capital, private property, and labor market. Therefore, it's not capitalism. It's not anarchism or communism either - it lacks flexibility of workers to do something else, or state planning telling workers to do something else. At best, it could be a model of a feudal society (where the "state servants" are the landed nobility). Furthermore, it has no concept of natural resources, energy, labor and goods. That has to be there before we even get to things like money. Also, the money definition is problematic. Money is essentially a contract, and can be created even between three private parties as resoldable bonds. This isn't there. I would also suggest, if the author (or anybody else) is interested, look at Steve Keen's work in economic simulation and his Minsky program. This guy is far ahead of everybody else in actually trying to understand how real economies work. To people who wonder why don't we have good economy simulators that would reflect a real world. Well, I think the answer is pretty simple - rich people don't want that. They don't want plebes to understand what it means to be rich, most rich people know they wealth is undeserved, and don't want other people to see it flaunted openly. So they prefer hegemony (in Gramsci terms) of neoliberal economics, which is misleading about what is happening in reality.

kamel3d 3y ago
Would rich people be able to stop you from creating a good economy simulator? At the end of the day, it's programming. Someone who has studied economics could make that without rich people coming after him. I argue that no one would come after you if you were to do so.
- js8 3y ago
  Yeah, but you also need to compare it with real world data. Without grounding, the simulation will be very limited. So you need to understand the real-world monetary flows to a sufficient level of detail. You need to understand who actually owns what, and how the power is exerted. And the rich people are against that, for example, in most developed countries (with IIRC notable exception of Norway and Sweden) the tax fillings are private. Also, it's kind of difficult to understand actual production, because private companies keep the data on production costs secret. On top of that, there is little culture of data sharing in economics profession. Again, mostly because the data are very valuable to companies and individuals, profit takes precedence to public understanding. So the fact of the matter is, lot of powerful people (and perhaps most middle class as well) don't want the level of transparency required to build a meaningful model. It's true that nobody will come after you if you try to program it. But also, you won't get much support from the existing, mostly neoliberal, institutions.
- wildrhythms 3y ago
  Yes because a rich person demands I pay them money for housing, which among other things I need to survive, and my precious programming time is spent under a different rich person to get that money.
01100011 3y ago
> rich people don't want that You had a great comment until you went off the rails. Sadly, economic ignorance is much more easily explained by apathy and the rarity of systems level thought in the general population.
- js8 3y ago
  That wouldn't be a problem, just look at climate science, it has decent models of very complicated systems, grounded in real-world understanding. But it requires lot of data to be collected and made available for people who do these simulations.
xpe 3y ago
> To people who wonder why don't we have good economy simulators that would reflect a real world. Well, I think the answer is pretty simple - rich people don't want that. The comment above is an example of an explanatory theory [1] > An explanatory analysis will try not only to describe the information but also to provide causal relationships between the various data presented. Simple, yes, but also too simplistic. It isn't very persuasive. For the theory to be useful, then "rich people not wanting that" would have to be causally relevant. The comment above only provides an implied incentive (e.g. rich people would benefit by preventing knowledge dissemination that undermines their status). However, there is no convincing argumentation showing how such incentives play out, much less that they are significant enough to be causally important. I don't want to reject the entirety of what the commentor above might be trying to say. There are probably aspects that are interesting, surprising, or maybe even troubling. Perhaps there have been documented cases of influential elites shaping economic theories and how they get communicated. Maybe; I haven't studied the history of bias in the field of economics per se. I know of many biases, but so far I've tended to think about them as being largely intellectual oversimplifications rather than designed misdirections. Let's look to some other 'simple' explanations. One competing theory would be that economists and modelers want to earn status and earn a living. There doesn't have to be any malicious nor coordinated action to explain such behavior. There are many levels of systemic behavior that align with powerful interests without any causal chain. This is sometimes referred to as co-evolution or co-adaptation. I do want to reject the whole family of (not so good) explanatory theories that fail to give sufficient argumentation for causality. [1] https://philosophy.stackexchange.com/questions/30827/what-is-the-difference-between-explanatory-descriptive-and-predictive-analysis https://philosophy.stackexchange.com/questions/30827/what-is...
- js8 3y ago
  I thought about your comment quite a lot. I don't think whether something is conscious act or not has nothing to do with causality. Steve Keen actually did some simulations according to which oligopolies can form without firms even knowing each other, just through a common pricing mechanism. That doesn't mean that oligopolies are not causally responsible for lower economic performance. You already seem to accept that there is evidence that powerful people do not want economic transparency. This is my (anecdotal) observation as well, some of it directly from economists. As far as economics as a profession, it's odd that (neo)classical economics held for a century despite many criticisms against it over the years (in particular the criticism that it is never really positivist). It seems pretty inline with the idea that powerful people are willing to pour a lot of money to maintain its hegemony. Whether the economists realize it.. Steve Keen discusses this a lot. Lot of neoclassical economists did some research themselves that disputed basic neoclassical assumptions, only for these assumptions to be continued to be taught to the next generation. So lot of them are probably pretty honest, just misguided.
tomcam 3y ago
> rich people don't want that Where does this sentiment come from? And what do you define as rich? I have been acquainted with rich people for decades now and many of them seem pretty darn open to me.
stevenjgarner 3y ago
So if you feel that way, why not program your own model introducing the "concept of capital, private property, and labor market"?
- xpe 3y ago
  > why not program your own model I wouldn't use the word 'program' here; I think it misses the essence of quantitative modeling. Quantitative models can be defined on purely mathematical terms. Some of these also benefit tremendously from computer simulation. Still, many can be analyzed and understood based on mathematical properties. The worlds of declarative programming and formal analysis can come together in very interesting ways w.r.t. modeling and simulation. P.S. This comment isn't aimed particularly at the commenter above; I'm writing this comment because I don't think the domain of 'modeling and simulation' is generally well-understood among software developers nor HN participants.
- js8 3y ago
  Honestly, I'd love to, it's on my list. I really like what Steve Keen is doing. But I am currently amusing myself with a math problem, and that's more important.

by js8 3y ago

kamel3d 3y ago

Would rich people be able to stop you from creating a good economy simulator? At the end of the day, it's programming. Someone who has studied economics could make that without rich people coming after him. I argue that no one would come after you if you were to do so.

js8 3y ago
Yeah, but you also need to compare it with real world data. Without grounding, the simulation will be very limited. So you need to understand the real-world monetary flows to a sufficient level of detail. You need to understand who actually owns what, and how the power is exerted. And the rich people are against that, for example, in most developed countries (with IIRC notable exception of Norway and Sweden) the tax fillings are private. Also, it's kind of difficult to understand actual production, because private companies keep the data on production costs secret. On top of that, there is little culture of data sharing in economics profession. Again, mostly because the data are very valuable to companies and individuals, profit takes precedence to public understanding. So the fact of the matter is, lot of powerful people (and perhaps most middle class as well) don't want the level of transparency required to build a meaningful model. It's true that nobody will come after you if you try to program it. But also, you won't get much support from the existing, mostly neoliberal, institutions.
wildrhythms 3y ago
Yes because a rich person demands I pay them money for housing, which among other things I need to survive, and my precious programming time is spent under a different rich person to get that money.

01100011 3y ago

> rich people don't want that You had a great comment until you went off the rails. Sadly, economic ignorance is much more easily explained by apathy and the rarity of systems level thought in the general population.

js8 3y ago
That wouldn't be a problem, just look at climate science, it has decent models of very complicated systems, grounded in real-world understanding. But it requires lot of data to be collected and made available for people who do these simulations.

xpe 3y ago

> To people who wonder why don't we have good economy simulators that would reflect a real world. Well, I think the answer is pretty simple - rich people don't want that. The comment above is an example of an explanatory theory [1] > An explanatory analysis will try not only to describe the information but also to provide causal relationships between the various data presented. Simple, yes, but also too simplistic. It isn't very persuasive. For the theory to be useful, then "rich people not wanting that" would have to be causally relevant. The comment above only provides an implied incentive (e.g. rich people would benefit by preventing knowledge dissemination that undermines their status). However, there is no convincing argumentation showing how such incentives play out, much less that they are significant enough to be causally important. I don't want to reject the entirety of what the commentor above might be trying to say. There are probably aspects that are interesting, surprising, or maybe even troubling. Perhaps there have been documented cases of influential elites shaping economic theories and how they get communicated. Maybe; I haven't studied the history of bias in the field of economics per se. I know of many biases, but so far I've tended to think about them as being largely intellectual oversimplifications rather than designed misdirections. Let's look to some other 'simple' explanations. One competing theory would be that economists and modelers want to earn status and earn a living. There doesn't have to be any malicious nor coordinated action to explain such behavior. There are many levels of systemic behavior that align with powerful interests without any causal chain. This is sometimes referred to as co-evolution or co-adaptation. I do want to reject the whole family of (not so good) explanatory theories that fail to give sufficient argumentation for causality. [1] https://philosophy.stackexchange.com/questions/30827/what-is-the-difference-between-explanatory-descriptive-and-predictive-analysis https://philosophy.stackexchange.com/questions/30827/what-is...

js8 3y ago
I thought about your comment quite a lot. I don't think whether something is conscious act or not has nothing to do with causality. Steve Keen actually did some simulations according to which oligopolies can form without firms even knowing each other, just through a common pricing mechanism. That doesn't mean that oligopolies are not causally responsible for lower economic performance. You already seem to accept that there is evidence that powerful people do not want economic transparency. This is my (anecdotal) observation as well, some of it directly from economists. As far as economics as a profession, it's odd that (neo)classical economics held for a century despite many criticisms against it over the years (in particular the criticism that it is never really positivist). It seems pretty inline with the idea that powerful people are willing to pour a lot of money to maintain its hegemony. Whether the economists realize it.. Steve Keen discusses this a lot. Lot of neoclassical economists did some research themselves that disputed basic neoclassical assumptions, only for these assumptions to be continued to be taught to the next generation. So lot of them are probably pretty honest, just misguided.

tomcam 3y ago

> rich people don't want that Where does this sentiment come from? And what do you define as rich? I have been acquainted with rich people for decades now and many of them seem pretty darn open to me.

stevenjgarner 3y ago

So if you feel that way, why not program your own model introducing the "concept of capital, private property, and labor market"?

xpe 3y ago
> why not program your own model I wouldn't use the word 'program' here; I think it misses the essence of quantitative modeling. Quantitative models can be defined on purely mathematical terms. Some of these also benefit tremendously from computer simulation. Still, many can be analyzed and understood based on mathematical properties. The worlds of declarative programming and formal analysis can come together in very interesting ways w.r.t. modeling and simulation. P.S. This comment isn't aimed particularly at the commenter above; I'm writing this comment because I don't think the domain of 'modeling and simulation' is generally well-understood among software developers nor HN participants.
js8 3y ago
Honestly, I'd love to, it's on my list. I really like what Steve Keen is doing. But I am currently amusing myself with a math problem, and that's more important.