A chilling mathematical model of inequality
sltrib.com
sltrib.com
Does this mean that the paper supports basic income? I don't know much about these topics but I see basic income pop up in conversation on HN every now and then, so it came to mind when I read this.
Cantillon effects: In Essai, Cantillon provided an advanced version of John Locke's quantity theory of money, focusing on relative inflation and the velocity of money.[61] Cantillon suggested that inflation occurs gradually and that the new supply of money has a localised effect on inflation, effectively originating the concept of non-neutral money.[62] Furthermore, he posited that the original recipients of new money enjoy higher standards of living at the expense of later recipients.[63] https://en.wikipedia.org/wiki/Richard_Cantillon#Monetary_the...
Right now, banks and then gov't are the first recipients of new money being injected into the system a la central banking.
Wealth distribution schemes implemented through fiscal policy (taxes and tax credits/grants/UBI) can be undermined through monetary policy this way.
Basic income proponents tend to focus on on sums big enough to live on though, but on this particular topic the size isn't really an issue, and smaller amounts could be interesting to try. Things like http://www.qe4people.eu tend to come up as a type of basic income. More permanent things like the Icelandic souverign money proposal ( https://www.forsaetisraduneyti.is/media/Skyrslur/monetary-re... ) has similar elements.
Providing the source is better than nothing, but we still have so far to go educating academics about the vital need for real and effortful data provenance and reproducibility.
But yes, I would recommend that academics check out http://zenodo.org/ which not only helps with the tagging of a github repo but also provides a citable DOI.
While we're at it, we might as well point out that source code alone is useless if it doesn't come with instructions on what kind of environment is required to install and run it, as well as the actual input data files used by the authors.
[1] Insult to injury: computer science, not physics.
Not a joke: courses supposed to teach programming are organized with Dropbox and email attachments.
Time and again, new services have come out, for decades, that supposedly will help solve the versioning, collaboration, and reproducibility issues with academic papers.
They never work, and the tools are always misused in this same way: as a catch-all bucket for disparate and unrelated snapshots of the paper, the code, and the data.
Merely moving from manual versioning in Dropbox to manual versioning in git is not in the least bit to be considered progress. We have to absolutely curb the underlying problematic behavior (e.g. this repo has but a single commit, the concept of version doesn't even apply) before it counts as even the tiniest step in the right direction.
In fact, I see it as a huge problem that people are chiming in to say that using git is in and of itself some kind of positive step. Why?
I don't care if you use svn, darcs, mercurial, git, perforce, or any other versioning system. The fundamental issue is actually keeping track of the changes you've made, and ensuring the precise version used for publication is trackable and available.
Even using Dropbox with clear documentation of changes made would be better than using a git repo where, prior to your first commit, you have destroyed all version history information and manually put everything into the "good" commit that you feel like "showing" to everyone.
That behavior is the entire problem. Using git, in and of itself, doesn't represent any degree of departure from the old practices. Just because git is popular and new-ish and sexy doesn't mean simply dropping stuff in a git repo constitutes progress. That's a dangerous implication.
Could you open an issue that could show them how to do what you're suggesting?
We can only take the authors' word for it that in between getting results for the paper and releasing the commit, they didn't make any significant alterations that would affect reproducibility.
The point is not to be critical of these particular authors. Good on them for sharing their code.
The point is to highlight that it's harmful for academics to "hide" their code until they get it all together for the "perfect" commit. It defeats the point of version control and their repo functions instead as basically nothing but a download link.
All the reproducibility problems would remain the same as if the code was provided as just a tarball linked from some academic web page.
Obsessions with code history are a little disconcerting. It's interesting to see, but has little relation to the quality of the paper (or the code base, for that matter).
Wanting to see code history is not anywhere close to an "obsession" -- that's just a loaded word that does nothing but distract from the issue.
> It's interesting to see, but has little relation to the quality of the paper (or the code base, for that matter).
This is just completely wrong. I don't want to see code history because I care about the author's earlier mistakes, previous implementation attempts, refactorings, etc., from any sort of judgemental, coding skill point of view.
The reason it's important is that maybe they used version 3 when they ran some part of the study and generated some of their data. They figured it was good enough, but then later they updated some code to version 4. They didn't need to re-run things for that older data, but they needed version 4 for something else.
But, version 4 makes a change that causes an error, or an undefined behavior, or a change to some statistical degrees of freedom calculation or something, that would imply the data from version 3 also needs updated.
So the authors are using something that came out of version 3, looked good, and they thought never needed revisiting. Then they are also using something from version 4 that, overlooked by them, invalidates or changes something from version 3.
All of this gets whitewashed because they go through and manually finalize the code, and upload it to a public repo.
Now, you as the reader are not even aware that there ever even was a version 3, let alone that some of the data came from version 3, some came from version 4, and the mixup is based on an oversight or something.
The code history really matters. If you cannot reproduce the results immediately with the tarball the authors release as the "good" version, and there is no version history, then, effectively, there is zero reproducibility. You have to push back on the authors, who weren't making their changes under version control when the version 3 to version 4 issue arose, and so good luck ever getting an answer.
I've actually experienced this first hand. I had to reproduce some quant finance research on country-based beta calculations. It turned out that in an early draft of the work, the team had used a completely different data set and in the earlier data set, the country code for Sweden had been set to "SW" even though the ISO country code for Sweden is "SE". They got some intermediate statistics out of that data set and saved them for later usage.
Then later on they used another data set and someone had the bright idea of using "SW" for Switzerland, even though that country code is "CH".
But because all that remained from the first data set was aggregate statistics, and no version history of the code that had loaded the spreadsheet and made these assumptions at that time, there was no way to figure this out.
In the final result, Swiss and Swedish data was being aggregated together and throwing numbers off. And those two countries both have very small market caps, so this is hard to spot and track down.
After a ton of outrageous debugging, we just happened to get lucky and notice that the difference was almost equal to the largest Swiss company in that time period (Nestle I think), and then began to wonder about the "SW" issue and then tried all permutations of separating the two data sets into Swiss and Swedish data until things matched.
If we could have bisected the original source revisions, it would have saved weeks worth of tedious, expensive data drilldown nonsense to replicate a result.
Besides, even if I did have an obsession with reading source code, it would not have any bearing on the usefulness of expecting version history in general. Here you are talking about me which makes pretty much zero sense, whether I am obsessive about it or not.
It's important to write a lot about this because the nature of your comments borders on unstable hostility. Even if you disagree with what I am saying about version history, and even if you disagree with how passionately I feel about it, I still deserve better than to endure petty insults that literally have nothing to do with the context of the conversation.
In science, reproducibility is what matters, not the history. Although if someone didn't tag the revision they used to generate data (or to compile and send to customers), that's pretty annoying.
Without the history to prove exactly what happened and when, there is no reproducibility.
I would even go further and say that each and every model fitting attempt should be recorded in some type of version history. This way, researchers can't engage in multiple comparison errors, whereby they test lots of things and fish for a specific threshold of significance that may be met solely by chance based on how many things they tried, or researcher degrees of freedom problems where seeing the output of intermediate model fits causes you to change the experimental design mid-stream (like collecting more data in the context of a model where that would change the required significance threshold).
These kinds of pitfalls are extremely common and go unreported most of the time. If someone found a significant relationship after testing hundreds of things and fine tuning the data collection, even if that result is statistically significant we should deeply discount it. But what if there is no record anywhere that all those previous attempts took place and that they motivated the experimenter to tweak the experimental design?
I enjoy thinking about economics because I am fascinated by systems. And economic systems are pretty cool because we all participate one way or another, they behave in ways that we might individually dislike (like "fair" market values) and as we are all components in a larger machine we get to try to see the "big" picture to understand what actions are available to the individual cogs.
And as the system arises as an artifact of our individual actions, it cannot "break", it can only "change". Further, when it does change it does so over months or years as peoples actions change and new equilibrium emerges.
Many, if not most, of the stories posted take some aspect of the system and extrapolate it out to some weird end point. My favorite example is, "Housing is so out of control in SF soon nobody will be able to afford to live there." On its face, extrapolations like that are silly. If nobody lived in SF, there would be no companies and no businesses and no need for city services, it would be a giant scrubland. Which absolutely nobody believes would ever happen. It doesn't happen because prices rise until nobody (individual action) is willing to buy at the price, at which point sellers have to lower their price (individual action) if they want to move away and realize the economic value of their home. Or more crisply, the system adapts to what is happening and that adaptation becomes the new normal.
So when I read these sorts of articles/stories, I always wonder "What are they really trying to say, the premise is clearly bogus so that can't be it." And in this particular case it seems to be trying to make people "care more" about inequality. Did you care more? If so why? If not why?
We don't need to play this game - there is plenty of room for both the wealthy and the not-wealthy. The more interesting question is what is it about the people who are already there that gives them the right to prevent development on land they don't even own, but simply look at.
What makes a diamond worth more than a ruby? What makes a Rembrandt painting worth more than my daughters Fathers Day card from the third grade? How is it fair that the cars get waterspots in a rainstorm?
The question you are asking is founded in that "wealthy" is a thing rather than it is an artifact of a behavior or event. And that question speaks to my basic point, why are the artifacts (and extrapolations) treated as if they were the important thing in the discussion?
Lets say you won the PowerBall lottery on Wednesday and elected to take the lump sum. $58 Millon, if you live in California you pay 54% in combined federal and state taxes so you clear $26 million dollars. You can now go into San Francisco and qualify to purchase any number of properties. Zillow claims there are over 1,000 homes available to you that you can now afford easily.
Let's say you graduated from college in 1980 and got a job in the Bay Area as a programmer and have lived in a 1 bedroom efficiency apartment since that time and saved all of your excess income in the Vanguard S&P500 mutual fund. Here 36 years later you have about $3.2M available and you can buy any of over 750 homes in San Francisco for cash.
You live in a small two bedroom apartment in Mountain View and your paternal grandfather dies. He left his estate to his grand children and after being distributed between you and your sister you receive $5M. You can now buy a home for cash in San Francisco.
In all of those cases you are now able to buy a house in San Francisco, did "deserving" ever enter in to it? Why or why not?
In economic terms the control of capital moves from actor to actor, sometimes that actor is required to relinquish control (for example you buy food to eat so you give money for food) and sometimes the actor chooses to relinquish control (you trade money for a nice new car) and sometimes the actor chooses to retain control (you loan it to others temporarily in exchange for them giving you back more capital than you loaned them, you do this by putting your money in a savings account, money market, mutual fund, or other investment). The amount of capital you get to control over your lifetime is a mix of luck and work. What you do with it once you get control is entirely up to you of course.
There isn't a social justice issue here, there is just the flow of capital amongst people who choose to use it where they will. It could be you, it could be someone else, but as much as it would be great if it was, the game isn't really is not rigged in the US, and that is doubly true in the Bay Area. Houses are expensive in San Francisco because a tremendous amount of capital is being sent into the economy to create new things. AirBnB raises $100M that is capital that comes in, they spend it on office space (and it flows to construction companies and construction workers) they spend it on engineers and data scientists and it flows to people who are probably reading this list, they spend it on advertising and it flows to advertising agencies or city coffers via bus stop advertising. Most of it flows into the control of people in the area. Some of those people will collect it until they can make competitive offers on a house, others will collect it and then move out of the area, and still others will spend everything they are given on new cars, vacations, and conspicuous consumption. But AirBnB exists because some actors (in this case investors) have moved capital into them, in the hope that later they will be able to exchange the equity in that investment for more capital than they originally sent over.
Are their social justice issues? Sure, but home prices isn't a social justice issue, its an economic artifact. Consider abandoned suburbs of Houston or Detroit where houses were abandoned because there was no capital coming into the city any more (Oil in Houston's case, Cars in Detroit's) It has been offered (but I cannot verify) that the greater metro area of Detroit had at one time housing for over a million people which was abandoned. Think about that when we discuss San Francisco's homeless population of 8,000 to 10,000 people (which is considered "large"). We tax the economy to use that capital to support the mentally ill, the less fortunate, that is a social justice issue. But it has nothing to do with home prices and everything to do with competent governance.
A more "charitable" interpretation might be that many long-time SF residents are fearful that if rents continue to rise, they (and those they associate with) will be forced to leave the city they love, and their former homes will be converted to office space and short-term rentals --- non-residential uses, hence "no one lives there".
Is there a scenario where this happens? Sure, it seems possible that tourists, businesses, and real estate speculators are willing pay more more for space in SF than long-term residents can or will. And as the city switches over to be less residential, the number of people willing to pay a premium for permanent residency may drop, accelerating the process.
The real question might be who benefits from preventing this "change" from happening. Someone who likes their current SF lifestyle would seem to have legitimate self-interest in preventing the city from changing in a way that would prevent them from continuing their lifestyle.
So at least in this case, I think people are really trying to say "We've got a good thing going here in SF, it's starting to move in a bad direction, let's not destroy what we have in an attempt to maximize economic gain." One can disagree with it, but I don't think it has a hidden agenda.
The paper says
The model consists of N agents, each with capital ci with i = 1, . . . , N. Agents are allowed to trade among themselves M objects. Each object m = 1, . . . , M has a price πm.
Starting from a feasible allocation matrix A, we introduce a random trading dynamics in which a good m is firstly picked up uniformly at random among all goods. Its owner then attempts to sell it to another agent i picked up uniformly at random among the other agents. If agent i has enough cash to buy the product m, that is if `i > πm, the transaction is successful and his/her cash decreases by πm while the cash of the seller increases by πm.
In other words:
1. Prices cannot vary. The price a good is set in stone.
2. Agents do not purchase things they want. They "pick up" goods and immediately try to sell them.
3. They sell them at these pre-determined prices to other people, whether the other people want them or not. Well, there's no such thing as "want" in this model.
4. Nor is there any planning by agents. No ability for them to say "Crap, I only have $5 left. Better not spend it on this vase this guy is selling."
Original discussion source:
http://www.econjobrumors.com/topic/physicists-model-economic...
1. The good is a commodity, i.e. its labeling has no effect on perceived value
2. The selling firms are perfectly competitive (i.e., no partial or complete monopolies that allow price-fixing)
3. The buyers are perfectly competitive (i.e., no purchasing groups)
4. All sellers can sell to all buyers equally easily
5. All parties in the market have equal access to information
6. All parties act rationally and based only on the marginal utility of each individual transaction
etc., etc., and a whole litany of other things that we know are rarely true. Since the theory is so "squishy" in practice, one could even protest that it does more harm than good when treated as a de facto starting point for price modeling: https://en.wikipedia.org/wiki/Supply_and_demand#Criticisms
So this model, even if it too is sadly oversimplified, may still be useful simply by starting from a different set of assumptions than whatever is typical in the field, and seeing how the modeled behavior differs from existing theories. There's basically no hope of experimental validation for any of this, though.
In another way, the same arguments could be made in a mathematical model and an essay, but the model might just be easier to criticise. I'd say that's a good thing.
But how those "scientists" deduce things on the real-world based on these simplistic models is utterly unjustifiable. In many cases, we know that some statistical properties show the models are inadequate, yet they still dare to write their BS about how "the internet" or "social networks" work based on such trivial models... (and the media disseminates these idiocies, like Ted and Wired...).
Sad!
As more market makers enter this situation, one of the main ways they compete is by lowering prices, which is good for consumers who rely on them for liquidity on demand. The market makers are essentially willing to put their money at risk by scrambling it around all day in the form of ownership of different assets, and then squaring up by the end of the day so that they hopefully don't hold any exposure to any asset over some significant time period.
This isn't about financial market making per se, but just the basic idea. There could be similar market makers in other fields, perhaps buying and selling physical goods, but never storing them for sale over time.
Fields where the handling and registration of goods transfers can't be as highly digitized as finance or energy would be much more expensive to provide liquidity, and the "bid-ask spread" would have to be higher to compensate.
In a sense, it would reflect the kinds of liquidity people care about buying, and if other kinds of liquidity dried up, it would be because people mostly didn't want to buy that liquidity.
I'm not saying that there wouldn't be unhealthy effects from inequality, but I think to the extent that inequality inherently limits liquidity, then market makers enter to provide that liquidity as a service.
This paper is basically a stone's throw away from the Marxian analysis of crises of capital. It's nice that it's quantitative; a nice complement to all the qualitative criticisms of capitalism's fundamental structure. It's not the market that is the issue, but the tendency of capital to concentrate in the hands of the propertied class. There are two kinds of capitalist societies actually seen in the real world, the unstable kinds, and the kinds that have significant wealth redistribution mechanisms. Real economies oscillate between the two or collapse.
On the other hand, there definitely are other forms of exchange, say like dark pools, where wealth and status are required to gain access.
I think I would hold out for a model that was able to work with actual transaction data like this, rather than needing the mathematical conveniences of the paper (that only budgets are limiting, that transaction network is the fully connected graph of all participants, etc).
It's good work, I hope to see it keep going and go in a more data-driven way.
[1] http://physicsoffinance.blogspot.com/
[2] http://physicsoffinance.blogspot.com/2016/02/economic-growth...
http://www.amazon.com/Ubiquity-Catastrophes-Happen-Mark-Buch...
http://www.amazon.com/Nexus-Worlds-Groundbreaking-Science-Ne...
A physical model of network effects that presents phase transitions gives us a view into the underlying reality. So cool.
Sadly, for 100+ years we've known about the destructive effects of deflation on leveraged economies (Much longer for the accelerated growth that comes with leverage), yet the leadership in political economics continues to call for hard money and budgetary surplus, which doesn't even make sense with respect to accounting identities.
> A physical model of network effects that presents phase transitions gives us a view into the underlying reality.
Here, imperfect mental models (aka opinions) are replaced by imperfect computational models. There's nothing inherently superior about this model just because it has been translated into code. It should be regarded as a tool to understand the consequences of the assumptions that went into its design. If those assumptions are unsound, the model is unsound and will not reflect real economic systems.
That's pretty simple, alright. Gee whiz. It's a good thing we have those abstruse theories from physics to tell us that people without money don't buy as much stuff.
I'm guessing this is a case of bad journalism -- anyone have a better layman's explanation?
I doubt that the answer would be 'one where everyone is exactly equal,' because we do have differing physical and intellectual capabilities which would mean that if everyone was exactly equal, we would be under-allocating resources to certain people. On the other hand, it's almost assuredly true that the level of inequality we have now is just too high. The difficulty is in finding the exact equation to maximize.
Rather than inequality, the problem is really that the bottom end of the range is too low. There are too many people who are not able to find enough work to provide an adequate income. There's a litany of social and economic reasons for that, but that is the problem, not the inequality itself.
The second network capital is concentrated and I borrow $100,000 dollar from two people (mediated by a bank) and buy a home.
Now I need to pay that capital back, with intererest. I can only do that by making shoes for people. Which network am I going to have a harder time doing that in? The one with 1000 people who have $200 saved in the bank? Or the one that has 2 people who have $100,000 in the bank?
Assume all people in the example have two feet.
Maybe that person is Bill Gates. Maybe it is the Koch brothers. But unless you strongly believe wealth correlates to competency and good will, it seems the system will inevitably attract abuse (power corrupts).
It may also be a problem for those other reasons, but that's not necessary for it to be a problem.
(People often mistake output and consumption measures of typical economic statistics for direct measures of experienced utilities. While they are proxies for utility, they are imperfect proxies, and one of the ways we know they are flawed is that they do not account for the disutility that directly results from inequality.)
Economists regularly use "Pareto Optimality" to solve for an optimal steady state or x-possibility frontier - a state/allocation where it is impossible to make any one person better off without making another worse off. Do you think this is a fair or useful way to determine the optimal allocation of resources?
If your solutions isn't Pareto optimal, that means we can make someone better off for Free.
Of course, just because something is Pareto optimal, doesn't mean it's the best solution. But the best solution is DEFINITELY going to be Pareto optimal.
> There is no optimum level of equality than can be solved for using a mathematical equation
This is not true: We can choose any function to maximize. As you correctly point out, the usefulness depends upon the validity of the quantity we're trying to maximize.
I don't know enough to say if Pareto efficiency is optimal here or not.
This is only true if one assumes private property and resource equality at the same time. If one abandons private property and tries to imagine systems of resource distribution based on democratic processes, there are far more possibilities. One possibility that obviously doesn't work out is a command economy where a centralized bureaucratic state owns everything. These systems seem to have similar dysfunctions to a megacorporation that owns everything in a country. We've never seen a decentralized, democratic system of communal property at large scale because they're always destroyed by violent means by other systems if they get too big.
A sprawling centralized state bureaucracy is none of those things. Indeed, the whole departure of Leninism from Marxism is the idea of a vanguard party transitioning from capitalism to communism via a state apparatus. It's a fundamentally elitist doctrine that quite obviously did not work the way it was presupposed to work.
It's not like people don't talk about capitalism in heavily idealized terms, either. I've never seen an exegesis of liberalism that doesn't elide the pervasive and frequent expropriation and violent enforcement required to prevent the system from falling over.
I'd love to see the level of a basic income used as the knob to use to kick-start an economy. Or some argument as to why this is a bad idea.
[1] http://www.paulgraham.com/ineq.html
Is it ok to have inequality if everyone is at least middle class? It seems ok, but I think the problem with this is the artificial barriers to breaking into the wealthy class. Lots of more capable people tend to be held down in lower classes while less capable people are allowed handle the wealth because they were born into it, know the right people, and know the culture.
Of course there are plenty of exceptions to this trend.
Interestingly enough, Ron Paul was one of the few saying this all along.
That's what tax cuts do, without the overhead of a large central bureaucratic program to manage it.
Even if you solve absolutely all the problems related to the level-playing-field and give everyone the same chance, you're still going to have some people who produce more value than others. I would say that is the root cause.
The real super admins are people like the current US Treasury Secretary who has the highest voting power on the board of governors of the IMF. This particular super admin made a fortune at Citi by betting on the global economic collapse of 2008 while hoarding cash in British tax havens like the Cayman Islands and was then promoted to his current administrative post.[1]
This is a (sub) prime example of a super admin purposefully crashing the system in order to gain more power; greatly increasing global inequality in the process. This is a cyclical use case which has been occurring for hundreds of years in the global paper-based markets.
[1] https://en.wikipedia.org/wiki/Jack_Lew#Between_Clinton_and_O...