Modelling a Basic Income with Python and Monte Carlo Simulation
chrisstucchio.com
chrisstucchio.com
Way to confirm stereotypes that programmers are abrasive and narrow minded. Writing code may be a good way to learn programming; that doesn't mean it's ideal for discussing economics. For one thing, even for an expert the code takes longer to read than a baldly stated list of assumptions. And not everyone is a Python expert.
> if I’m wrong, either Python is computing the wrong answer, I got really unlucky [...] or one of my assumptions is wrong.
Umm no. This is a false alternative. How about a lot of assumptions being left out altogether. E.g. it's pretty clearly assumed that this doesn't affect economic growth, any benefits resulting from changed income inequality [1] are left out, inflationary effects do not exist etc. Any half serious student of economic modeling would have a field day picking out all the details routinely discussed in macroeconomics that this analysis is ignorant of.
And anyway who are these financial results even calculated for? If it's the government, then the expenses are fully incurred, but the "JK Rowling factor" is taxed at less than 100% to compensate. If it's the average person, you need to somehow factor in the benefit of reading Harry Potter!
Yes, and you didn't even mention some of the biggest assumptions. For example, because Basic Income is something that you get even if you work, you'd be looking at a completely different salary structure. If you earn $2500 now and a Basic Income of $1500 were introduced, then you wouldn't suddenly earn $4000. You'd earn something like $1000 (plus or minus a factor due to the incidental massive changes to the economy from a Basic Income/Basic Job setup) on top of your Basic Income. That also means that taxation would have to work completely differently.
How that turns out is anybody's guess. With an emphasis on "guess". Code cannot magically substitute for the lack of the completeness of the underlying model, or for the research necessary to figure out the model. If you make stuff up from whole cloth and then bake it into code, it still remains stuff that's been made up from whole cloth. Running a program does not validate the underlying model.
Yes, you would. That's what basic income is. If you make $1,000,000 per year and basic income of $10,000 is introduced, you now make $1,010,000 per year (before taxes).
>You'd earn something like $1000 (plus or minus a factor due to the incidental massive changes to the economy from a Basic Income/Basic Job setup) on top of your Basic Income.
This is similar to a completely different proposal called "guaranteed income" [1]. It's not what is being discussed by the OP.
[1] http://en.wikipedia.org/wiki/Basic_income#Difference_between...
You're misunderstanding me here. I'm saying the exact opposite of what you think I'm saying. My point is that the market for wages would find a new, lower equilibrium to compensate for the new basic income. Since there's only a finite amount of money in the economy, and we can't magically give everybody that much more, it has to come from somewhere. A Basic Income approach would still have to redistribute the existing money, not create additional money (at least not on such a scale). The average total income (basic income + wages) would remain largely unchanged (modulo changes in tax structure, GDP growth or shrinkage, etc.).
I'm not sure of your definition of money here- the Government prints currency all the time. If you mean goods and services, why can't we produce more? Are you saying that our current economic output is at it's maximum capacity?
So, you did. I missed that. So I guess my next question is why can't we just create more? The (increased) amount we would have to create would not be equal to the amount of the basic income, because we could eliminate most welfare programs and their associated overhead. On the surface, I don't see a problem with printing the rest.
Personally, I don't believe that the impact is going to be quite as zero-sum. As recent stimulus attempts have shown, when you give poor people cash, they happily spend it, contributing to the economy. This in turn helps support greater production of goods and services. If the production capacity were fixed, then prices would jump. Instead, we are likely to see production jump more.
This is my personal belief. Anyone that claims to actually KNOW what will happen is fooling themselves. From my understanding, NO ONE has tried this at the scale being discussed.
Closest I think is Alaska Permanent Fund, but it's very small individual amounts (highest was 2008 in the amount of $2,069 for the year). Source: http://en.wikipedia.org/wiki/Alaska_Permanent_Fund
Compare that with $14,000-$30,000 being discussed and you see that we have no idea what such an infusion will do. It might usher in a new golden age of arts and sciences, or it might make us all lazy and inept. Personally, I'd rather give it a go, than sit and guess forever. Ending poverty forever just sounds too good to pass up.
The article addresses what results are being calculated, and why other things you might wish to model are not included: "I’m going to focus on one specific aspect of it - a Cost/Benefit analysis studying the effects on the government budget, and [minus] the value of the work produced by people benefitting from the program. There are other aspects to consider, but I’m more interested in Monte Carlo than in politics, so I’m going to ignore those for the time being."
The ideology I'm pushing hard here is that you should carefully think and model things, rather than just reflexively saying "ooh complex, here are liberal/conservative talking points." I just picked on Basic Income specifically since I see a lot of discussions about it here (e.g. yesterday), and those discussions are just a mix of vapid and innumerate political talking points by people who clearly don't even have a mental model of the world.
If you want to build a better model, please do it. It will be vastly more useful than the normal discussion we see here on the topic of BI (and on many other such topics).
http://apps.npr.org/unfit-for-work/
without knowing how many would switch from disability to BI and/or BJ - hard to say (I would say based on the disabilities described, like severe back pain so no more construction, more would be inclined to switch to BI and not BJ)
I agree and that's basically my criticism of the article. The underlying economic model can be expressed in 2-3 arithmetic equations. It competes in a field where people routinely spend years to research very sophisticated models that nonetheless often prove inaccurate in the field [1].
To give an analogy, how would this sound as an argument to develop a business: "I assume that we can borrow 10M at 5% interest and pay it back over 5 years and that sales will increase 50% over that period. Here, I wrote some code and it turns out positive. I'm not interested in discussing the reasons for my sales forecast; if you have a different opinion, just write your own code." And we're not even talking about a business here, but the whole economy.
[1] http://en.wikipedia.org/wiki/Econometrics#Limitations_and_cr...
If you showed me your business forecast, I would indeed tweak the model until I felt the probability distributions governing the assumptions covered all the plausible values. The net result would likely just be a very fat distribution that said nothing. And if so, that's useful information to have.
That's very nearly the result we get here - note that the Basic Job estimates range from $300B to $1.5T. See Scarmig's excellent comment which cuts to the heart of this matter.
> That's very nearly the result we get here - note that the Basic Job estimates range from $300B to $1.5T.
The problem I have with this post is this. It is NOT useful information to have.
What is that range based on? Nothing. Your guesses. There is no known probability of the real-life result lying between those values, therefore the values are meaningless.
I'm all for models, models are great, but they need data. There's no point in a model based on assumptions, it's basically just a smokescreen for making your opinion look like fact.
> I would indeed tweak the model until I felt the probability distributions governing the assumptions covered all the plausible values.
Basically you would tweak the model until the result matched your opinion. If you had data of 100 similar startups and you calculated the distributions of their loans and sales growths, THEN AND ONLY THEN is your model useful
edit: I'm giving up on this thread now because I feel like this - http://xkcd.com/386/
So your call for programming as a response says more about how well you understand the problem space, than anything about an alternative to current political discourse. Not only will people never use mathematical models to express their opinions, even if they did they would still talk past each other and hold tightly only to the answers which confirm their own biases. You misunderstand why and how people exhibit this behavior if you think a numerical mode of expression is the solution.
I'm having a hard time thinking of a topic involving economics where the latter would be preferable.
From the post:
> If you disagree with these numbers or this distribution, there is no need for you to stop reading. Just download the code and change the numbers. Your change in these assumptions might completely invalidate my conclusions, or they might not matter at all. The best way to find out is to actually run the model.
Varying parameters and producing different predictions would not invalidate the model. What validates or invalidates a model is whether or not it actually predicts the behavior you're trying to model. Without a validity test, modeling is pure sophistry.
So instead of suggesting another naive model, let me suggest a way to augment the whole project: construct a model that successfully predicts actual, ongoing, real world behavior. Test your model against actual, real world policy changes. If it has verified predictive power, then you have some base to claim that it can predict novel policy changes.
Any given person's opinion may condense decades of personal experience. I disagree with the author's claim that (even naive) mathematical modeling is better than personal opinion. Given any rigorous definition of better, that is frankly an empirical claim that the author has left unsupported.
There are dozens of completely arbitrary assumptions and nobody could possibly know if they were anything close to being right
> If you disagree, write code.
Ok..
"With a Basic Job, you can’t - your choices are a Basic Job for $14480 or possibly a McJob for marginally more (say $15000). I’ll model this by assuming the number of non workers will increase by up to 5%, or decrease by up to 20%:"
I disagree, I made my model assuming that the number of non-workers increases by between 60 and 80%
Oh look! My results show that basic income is better! Great, that's what I believed before hand, my model must be good.
edit: I'm all for mathematical modelling, but there needs to be some kind of basis for the assumptions, and it helps if there are less than a bajillion assumptions too. Anyone could tweak all the assumptions so they still sound reasonable and come up with any conclusion they wanted from this model
Even with what I consider to be a wildly unrealistic assumption, the Basic Job is still cheaper than the Basic Income in well over 50% of possible worlds.
I repeat: write some fucking code. Don't just say you did, actually do it.
A set of iPython notebooks on Github could inspire the inner economist in us all.
> The basis of any informed discussion is a mathematical model. The best way to think of a mathematical model is a way to force everyone to clearly enumerate all assumptions being made, and to accept all logical reasoning that follows from those assumptions. Given a model, everyone involved in a discussion can agree that either the conclusions of the model are correct, or one of the assumptions going into the model must be false. This is very important when disagreement is reached - disagreement in the conclusions implies disagreement with some assumption, so it makes sense to figure out which assumption is the cause of disagreement.
It’s also important not to be blinded by a model. The involvement of numbers does not make an argument empirically correct - it simply makes it more understandable and less likely to be logically flawed.
He knows the limitations of mathematical modeling and is simply presenting a basic simulation of these two policies subject to various assumptions.
And what’s worse, they seem like something dependable. Isn’t it very hard to judge how much a model has to do with reality? A discussion about a model might be a perfectly scientific, constructive and relevant discussion about a theoretical world that has almost nothing to do with the real one.
The "real world" is a necessary component for the scientific method. By comparing model predictions to real-world experiments, it allows refuting and refining models and theories. How else would you evaluate theories, to separate the chaff from the grain?
The feedback from observation is absolutely vital.
You're basically saying that economy is not science, in the sense above. Which I agree with. It is madness.
This doesn't mean that things outside science cannot be useful -- just that it's not science.
"it simply makes it more understandable and less likely to be logically flawed" -- yes to the first part, no to the second.
I think Climate Change and Securitized Credit crisis also taught this lesson for me: people with models can persuade people into believing anything. It's just a new way to lie with statistics.
E.g. "I’ll model this by assuming the number of non workers will increase by up to 5%, or decrease by up to 20%: I disagree, I made my model assuming that the number of non-workers increases by between 60 and 80%."
Okay, now instead of arguing about basic income in the abstract, you're arguing about projections of how much the number of non-workers changes. The latter is a lot more concrete and easier to analyze further.
I really don't understand all the negativity here.
If you're modeling investment growth and your model is:
future_value = investment_amt * (interest_rate ^ num_years)
future_value = investment_amt * (1 + (interest_rate * num_years))
future_value = investment_amt * ((1 + interest_rate) ^ num_years)So, despite the fact that there was some trifling value to the post, he's really going to garner a lot of negativity.
"shut the fuck up and write some fucking code"
This is explicitly an assertion that there won't be anything important that can be added to the debate that cannot be expressed in code.
On the flip side, he absolutely doesn't assert that the particular model is the end-all in terms of anything - he expresses desire to see improvements.
Obviously gathering data would be an incredibly valuable contribution to the discussion. I should have been more clear on that point, I guess I just thought it went without saying.
> The basis of any informed discussion is a mathematical model.
And then towards the end:
> My conclusions are simply the logical result of my assumptions plus basic math - if I’m wrong, either Python is computing the wrong answer, I got really unlucky in all 32,768 simulation runs, or one of my assumptions is wrong.
> My assumption being wrong is the most likely possibility. Luckily, this is a problem that is solvable via code.
That sounds to me like him asserting mathematical models, expressed in code are the only reasonable means we have for knowledge. I don't think he's quite made that case.
Statistics don't come after lies and damn lies accidentally. A naive model of liquid cooling ends in a fatal explosion and the discovery of phase change.
(Disappointed by the downvotes... I'm not trying to be pithy.)
Incidentally, I am not convinced that tarot cards cannot be useful at generating insights - privileging random hypotheses and reconsidering your situation could be a useful means of reducing the impact of hypotheses you're privileging for other reasons. Everything must still be evaluated in light of actual evidence, though, of course, before conclusions are drawn.
Only if the model is accurate. If (as in the case of tarot reading and failed gambling) you have a system that doesn't correspond to the world you will draw conclusions that are detailed and reproducible but not insightful of the world. Inaccurate models are only misleading, not vaguely insightful.
https://gist.github.com/anonymous/7452196
Run it, and it will look very similar to what the more complicated model results in.
Almost the entirety of his model's result comes from the fact that the BI is given to everyone, while the basic job is not. All the other parameters he brings up are interesting asides but irrelevant to the actual results of the simulations. His model of the basic jobs proposal is pretty much equivalent to a proposal that "if you're unemployed, you should get a basic income; if you're employed, you shouldn't."
Basic income advocates consciously reject this. I don't have time to finesse the simple model right now to represent why we reject that proposal (I have, well, a job to go to =). But I'll hopefully get to it this evening.
Thank you. I look forward to seeing your alternative model. Please email me if this thread vanishes before you write it up, I really want to see it.
He had several arguments for this. One is, that at the moment in Germany, labor is taxed, so there is an incentive to move production to overseas. A VAT is applied to goods from abroad and from the same country fairly.
The basic income would then be a negative tax refund to compensate for the social problems that a high VAT causes (since poor people suffer under high VAT more than rich people).
So let us say a person spends 1000 EUR on VAT-relevant purchases (food, clothes, phone, etc.). Now a VAT is introduced, so the products now cost 2000 EUR. Then a basic income of 1000 EUR would annihilate the additional burden. Another person spending 2000 EUR a month at the moment would soon have to spend 4000 EUR, 1000 EUR being provided by the basic income, therefore this person would have to spend 3000 EUR (1000 EUR increase) to buy the same goods as before. Because other taxes would be abolished, the income would be higher. Furthermore, with VAT there would be less tax loopholes and tax collection would be a lot easier.
As a _second_ idea, this basic income could be chosen to be above the level necessary to compensate for the VAT. This is the idea which is in generally associated with the basic income and outlined in the model.
I will take a look at the model more closely. I find this approach really interesting.
While talking about basic income we should not forget that he's running one of the largest drugstore chains in Germany. If everyone would have a basic income the difference in income would be going into consuming products. So retail would benefit a lot from it.
On the other hand removing tax on labor would benefit retail too because it's a major cost factor there.
I'm not convinced that we would have better society if we would implement his proposed scheme.
An additional cost is the incentive or disincentive for engaging in productive work created by the Basic Income. On the one hand, the BI creates disincentives for recipients to engage in productive labor. If you are happy earning direct_cost each year, why bother putting down Call of Duty and getting a boring job? (…) My inclination is that it is more likely to reduce labor than increase it. So I’ll assume that the number of non-working adults will increase by up to 10%, or decrease by up to 5%.
This is always going to be the crucial input parameter that makes or breaks the whole idea, and no amount of math modelling is going to tell you how millions of people will behave after such radical change.
_My_ feeling is that people would always like a little bit more money. Basic income removes the "cliff" effect, whereby earning more money can actually reduce your income. Therefore people will work more.
After all, if you could go out and work for five hours next week, and be guaranteed that those five hours would give you the cash to buy something you have your eye on, then why wouldn't you?
The question is - once you start accumulating, how much will people prioritise "getting more stuff" over "playing computer games all day"?
I mean, I earn about $60k. I'm not about to drop down to a $10k income, even if I could get that without any effort. And I think most people are like that - a big reason, as far as I can tell, that many people who could work, don't is that they would lose money by doing so, at least to start with, and they're terrified to take that risk.
Let's say you're unemployed now - someone asks you to help them wash windows for a few hours next week. Doing so would jeopardise your benefits, and possibly leave you homeless if you were found out. If you could take that bit of work, you'd probably find the number of hours going up (if you were any good at it) and possibly leading into more, better, work.
But if we stop people even taking that first step, we effectively shut them out of the job market.
(I'm sure there would be _some_ people that decided to play games all day. But I'm fairly sure the free riders would be a tiny part of the population. People like stuff too much for that.)
I don't.
And that's the problem with the whole discussion: a simple, critical tipping point makes or breaks the whole proposition, and nobody can sufficiently prove which way it will tip to the satisfaction & persuasion of the sociopolitical opposition. Doesn't help that the proposition, hinging on this vapid presumption (either way), involves ENORMOUS amounts of money: "oops, we were wrong, too many people became free riders" would prove disastrous. Humans have tried a lot of socioeconomic structures, and a "play if you want to" BI model hasn't taken hold.
"Only two segments of Dauphin's labour force worked less as a result of Mincome—new mothers and teenagers. Mothers with newborns stopped working because they wanted to stay at home longer with their babies. And teenagers worked less because they weren't under as much pressure to support their families.
The end result was that they spent more time at school and more teenagers graduated. Those who continued to work were given more opportunities to choose what type of work they did."
Which isn't at all conclusive. One town, in one culture, does not mean very much in the long term. But it's enough for me to think that more experimentation is worthwhile.
Here is the code. You are interested in line 17.
from scipy.stats import *
Try installing a scientific python distribution like Enthought Canopy or Anaconda. On Unix systems it's possible to build it yourself, but on Windows unless you have the right compiler licenses and enjoy pain, you probably don't want to do that.
http://www.mirror.co.uk/news/uk-news/meet-the-man-who-has-10...
A basic job tends to discourage mobility in all the ways a normal job does. A basic income encourages mobility because you have time to do anything... including invent something or go back to school (or use online courses) to learn a totally new field. Or even just take a break to recharge.
I'd love to model this, but I have no clue how. Perhaps a normal distribution of a change in a person's ability to produce value when taking basic income and not working. Some people will drastically decrease their ability to produce value, some people will drastically increase it. Most, I suspect, will somewhat increase it and then get back to work.
The article didn't "ignore this completely". It set up a framework within which you can easily express that argument. Go do it.
There's absolutely no data to support any numbers for this, how could I possibly guess?
That's why I don't like this post. It would have been really interesting if there had been any basis at all for the numbers, but a model for something this complicated backed by nothing is worse than useless - it's misleading.
Stucchio didn't propose a magic python script that answers the question of whether we should a basic income or basic job. He simply proposed a tool for making the debate more concrete. The debate still happens. It just gets less stupid.
No. it makes the debate less concrete.
The model is absolutely 100% meaningless. An equally valid model would be:
I assume that basic job has a cost of $1 trillion (+/- $500 million) and basic income has a cost of $3.25-3.5 trillion
How does that improve any debate?
This whole thread is really annoying me because it's just hiding bullshit behind "maths". There are no insights, no conclusions, nothing. Just some numbers. Most annoying of all is the stupid "write some fucking code!" counter-argument. No. I wont write code with bullshit numbers to try and convince people of something that has no data to support it
People are going to go away from reading this article actually convinced that basic job is better than basic income based on what is essentially this guys uninformed opinion. That is bad
But there have already been experiments, like this one: http://www.dominionpaper.ca/articles/4100
"Only two segments of Dauphin's labour force worked less as a result of Mincome—new mothers and teenagers. Mothers with newborns stopped working because they wanted to stay at home longer with their babies. And teenagers worked less because they weren't under as much pressure to support their families.
The end result was that they spent more time at school and more teenagers graduated. Those who continued to work were given more opportunities to choose what type of work they did."
I think that this demonstrates to a large degree the author's point that the best way to disagree is to make a change and see what you come up with. For someone with inclinations similar to yours, your comment is likely enough for them to justify dismissing the result reached by the basic model, but when you add the proposed effect to the model, well, you come to the same basic conclusion, that Basic Income is significantly more expensive than Basic Job.
A lot of comments here, including yours, come from an unjustified implicit rejection of the main thesis of the article by preferring rhetoric over model refinement. That is, the author is arguing that a better way to go about this discourse is to test the effects of different assumptions with the model, but many commenters are hammering through rhetorical rebuttals to the model's conclusion without addressing whether they actually believe that rhetoric is more effective than the mode of discourse proposed by the author.
Your rhetorical argument (including the argument of the grandparent since you apparently concurred: "absolutely") suggests that the main factor in deciding between the two approaches is whether they decrease or increase work productivity. When that hypothesis is tested in the model, it turns out that the other factors combined overwhelm the effect on worker productivity, and so the effect on worker productivity should not dominate the debate as suggested by the grandparent.
This was an easily testable hypothesis given the model offered, and once it was tested it looked weaker than suggested by the grandparent. I think that this demonstrates that there is some truth to the author's claim that running our assumptions through the model, and refining it overtime is likely more valuable than making these untested rhetorical arguments.
Now, certainly there are likely problems elsewhere in the model, and reasonable people can have disagreements on what assumptions are reasonable. But the point is to argue and test those assumptions continually, rather than offering unweighted rhetorical arguments and either never reaching consensus, or just landing on the argument that is most repeated or shouted the most loudly.
If you _give_ everyone a job, then what are they doing? How does that compete with people already in minimum wage jobs? Are we offering, effectively, subsidised labour for large corporations? How does that effect people already in those jobs, or people who would be moving into those jobs?
I don't believe that any model could be made which would take all of the factors into consideration - I believe that we will need to _do_ it (in one city, or one state, or one nation) and then see what the results are.
You seem to be led by your conclusion that the issue is intractable and letting small complications (or misunderstanding of the proposal) defeat serious reflection on the issue at hand. Another benefit of the approach suggested by the author is to allow a framework for thinking through the problem. Policy has to happen, it's better to have a framework for thinking through the implications of policy changes than to throw your hands up in frustration because rhetoric is a wholly useless tool for making policy decisions.
Note that I changed the plausibility range of this parameter quite a bit in the "How to Disagree" section, it didn't alter the outcome much.
"It seems to me that this failure of the economists to guide policy more successfully is closely connected with their propensity to imitate as closely as possible the procedures of the brilliantly successful physical sciences - an attempt which in our field may lead to outright error. It is an approach which has come to be described as the "scientistic" attitude - an attitude which, as I defined it some thirty years ago, "is decidedly unscientific in the true sense of the word, since it involves a mechanical and uncritical application of habits of thought to fields different from those in which they have been formed.""
"Unlike the position that exists in the physical sciences, in economics and other disciplines that deal with essentially complex phenomena, the aspects of the events to be accounted for about which we can get quantitative data are necessarily limited and may not include the important ones. While in the physical sciences it is generally assumed, probably with good reason, that any important factor which determines the observed events will itself be directly observable and measurable, in the study of such complex phenomena as the market, which depend on the actions of many individuals, all the circumstances which will determine the outcome of a process, for reasons which I shall explain later, will hardly ever be fully known or measurable."
I hope some country/state out there tries out an experiment with basic income (just not mine). Politically, I don't think it will happen. A few poor people getting benefits cuts will ring a lot more loudly than middle class and rich people getting benefits raises.
16% of US government spending is on pensions (by the posters own source). Those people need money to live. Unless you expect them to work until they die / become disabled, then they are not covered by the naive model of a basic job system.
Indeed the basic job model completely ignores 52 million people who are not in the workforce or disabled (by their own numbers). Many of these people require government money to live as well. It is worth noting that in (small) trials in canada it was overwhelmingly students and parents that stopped working (and that is out of the labour force) and in much smaller numbers than the author suggests (although admittedly this may be because they knew the program was temporary).
Most of the costs and benefits involved in a welfare system are not present in this model. The suggestion that the only value people receiving welfare create is the occasional one in ten million great work is frankly shockingly old fashioned. It is as if 300+ years of political philosophy have simply passed the author by. To say the least you need to think about the effects on child care, students, crime rates and all the latent costs that a lack of financial freedom creates.
No, writing code is not the point here. Anyone can write code to do this kind of modeling. Reducing the huge amount of degrees of freedom in the model in a meaningful and fact-based way is key. And here the author fails big time.
The author does kindly gives us permission to critique as long as we accept his model - a very naive device with zero academic or economic justification.
Expert opinions certainly beat naive code.
This is not to say that code cannot be valuable, but the justification for this code does not even begin to touch upon real macroeconomic issues.
One of the possible concerns of a basic income is rampant inflation - it's clear that the author has not even considered this.
Peer-reviewed economic models would have to justify their model assumptions in a more robust way, including empirical data where appropriate. I'm not an expert but I expect this is a good place to start: http://www.journals.elsevier.com/economic-modelling/
All models use simplified assumptions and questionably appropriate techniques, I agree - that is practically the definition of 'model'. However, a useful model should justify those assumptions and techniques using current research-level knowledge, and/or show how the model applies successfully to reality.
I see this model to be as useful to economic discussion as saying "the world is flat, plus or minus 5 degrees". It ignores all current work in the field.
If you just say "oh the world is so complicated and I like basic income because maybe it will reduce the # of prisoners and reduce income inequality", I have no idea what you believe will actually happen or why, and you probably don't either. For a great example of this, take a look at this post: https://news.ycombinator.com/item?id=6725282
However, I am opinionated enough to have an opinion:
- I believe that basic income allows people to invest in themselves through education or simply taking small risks without the constant and desperate need to either work or degradingly grovel to the state to pay for their basic upkeep. This would allow people the option of taking entrepreneurial risks which are generally available only to the upper-middle- and rich-classes. I do not believe that everyone would take any such opportunity but I believe we should seek to give this type of opportunity to everyone, and a basic income seems a fair way of doing that.
- I believe that basic income would create an economic shock in any country that implemented it universally, the consequences of which are not immediately obvious, may cause difficulties perhaps including rent-seeking behaviour from certain sectors and gross inflation of necessities. I believe that the unknown consequences make implementation politically difficult, but that the eventual outcome would be good for society as a whole.
Can you add these to your model?
I don't know what your second bullet point is talking about - unlike a mathematical model, your words are unclear.
Given that, I would argue that my words are a better model for reality than mathematics :)
> you probably want to alter that until it describes your beliefs rather than mine.
Exactly. You are just stating your beliefs. The fact that you have stated them in a faux-mathematical way doesn't make them any more sensible or coherent.
Did you figure that out from the part of the blog post where I explicitly said "the involvement of numbers does not make an argument empirically correct..."?
The math simply makes my assumptions obvious and my reasoning clear. That's all it is meant to do. And again, if you believe my assumptions are not sensible, replace them with your own and rerun the model.
After you do that, I won't be confused about what you believe.
And again, I won't be doing that because I think representing the situation as a naive, untestable model adds zero value to any debate about Basic Income. I don't believe that what I said was especially hard to understand for someone who had an interest, and you could have asked me to clarify any point you wanted.
If this model is supposed to represent the total of your beliefs about Basic Income, then it's not clear to me that you have considered the issue at all.
The part where you say "I’m more interested in Monte Carlo than in politics" makes this explicitly clear.
Love this.
Also nice application of Monte Carlo simulations. They're great for thought experiments.
Even with a very simplistic model, there are too many assumptions that can be contested forever and that will ultimately only depend on what the people debating already think about it. I, for example, would make the assumption that the marginal value produced by a Basic Job worker in an hour would be close to 0, while administrative costs are underestimated by an order of magnitude. Now, it might be fun to try these out when I get home, but I don't think I can get any significant insight from it.
And if you were to refine the model, it would then become too complicated to discuss in ways that could be comprehended by most people - too much effort and they'll stop listening.
Moreover, it fails to account for larger benefits to society, which cannot be measured in the way such a simulation works. For this one we could seem crime dropping (happier society, prison costs diminishing, etc.), global health indicators improving (thus preventing tremendous cost), etc.
Lets focus on the latter. How many people commit crimes due to lack of a Basic Income? (If you don't know precisely, choose a probability distribution that includes all values you find plausible. For example, choose a uniform distribution on [0, # of criminals/year].)
How much do each of these criminals cost society? (If you don't know precisely, choose a probability distribution that includes all values you find plausible.)
Multiply the former by the latter and subtract it from the result. Rerun the simulation. Maybe the result will change a lot, maybe it won't change much at all. I don't know what will happen, but the simulation will tell you what could happen.
I do agree his model is probably overly simplistic. What could be awesome though is open sourcing something like this and having the debate evolve as a series of pull requests and forks that represent different assumptions and scenarios.
> The marginal income from labor under a BI is precisely (1 - tax_rate) * income, rather than (1 - tax_rate - lost_benefits) * income.
shuldn't the second one be (1 - tax_rate) * income - lost_benefits? I don't know if/how benefits are taxed in the US.
I should also point out that governments already create far too many "busy work" jobs as it is. The IRS and welfare systems are some of the best examples due to overly complex regulations and means testing.
I don't purport to know the answer which is why that number is a random variable.
There's no reason to imply that whatever the government tells them to do for their basic job produces more value than whatever they decide to do. It's pretty well established that the more government tells people what to do for work, the less an economy produces.
Lets make people pick trash instead of helping educate their kids, surely picking trash is of much more societal value...
How come there's no basic job attached to the money GS gets? How come no one is worried that rich people are going to stop working after we spent 5 years giving them all the money we can borrow?
How many stockbrokers do you know that quit their jobs because the government is flooding the market with free money?
HSBC didn't stop laundering money for drug cartels just because the US gov't was offering them all the free money they would ever want.
A policy like this could reasonably be said to cost nothing, produce nothing, and indeed do nothing. If we think that congress generally makes the world a worse place by implementing policies that do things, then it would be an improvement for congress to work on this sort of thing instead. However, its cost-benefit under the proposed model is not $0 or (pay_per_lawmaker * n_lawmakers * nop_debate_time) but $13 quadrillion. The proposed model indicates that this program costs about 180 years of global production at the current rate, or a substantial portion of the value of the entire planet and all humans living on it. If we suppose the planet is worth about $200 quadrillion, we will run out of it about 15 years after implementing this policy.
I think that might be the most interesting way of looking at such an abstract model.
The monitoring band would also act as the recipient's communicator (smartphone) - so - they can freely use it to call, text, email, get apps, etc., it's just that all the data they generate - each individual participant's trace - is recorded.
The massive amount of data collected from people willing to trade a homeless lifestyle for one with a steady income would have tremendous value in aggregate, and be unavailable to anyone for nefarious purposes. The only thing you have to do to get the basic income is be willing to have all your data recorded (but not shared publicly!)
Your anonymized data will then be made available (along with everyone else's) via an API to developers of apps, and, assuming you give them permission, they will be able to access your sensitive / personal / private data as well.
If the plan is to replace certain spending with a basic job or income, it would be interesting to see the comparative costs and efficiency of what we're replacing as well.
it does not change the comparsion of BI vs BJ but it is interesting to note that even with his assumptions it means the effective impact on USG Budget is approx 0 for BI and a big reduction of $ 2 trillion for BJ.
unstated assumption: and do so quickly enough for it to influence the discussion before it slips off HN
Really this article should probably be something along the lines of, "Monte Carlo Simulations in Python."
This is not a discussion of BI at all.
the task of building a model of labour market and estimating your model with real data is very complicated.
import random
def sky_is_green():
return random.choice([True])
A minor mistake. The only "sorta well known to Americans" basic income is the Alaska fund which does only pay to adults.
However most proposals seem to apply to "living humans" or whatever. After all, kids need to eat too. Or if using the franchise fee model of justification, kids lives are being ruined by the economic system too.
So, no, I have no idea where he gets the idea only adults would get an income.
A slumlord can TRY to raise his rent 10% anytime he wants, including today, but if the residents aren't making any more money, that plan is not going to work very well.
At the highest levels multimillion dollar lakefront homes aren't going to be affected unless we unleash subprime loans again (which would be profitable, so we will, but thats another discussion).
The primary effect would be on the shrinking middle class. Many of those extremely expensive welfare administrators are middle class and replacing those jobs with an extremely small shell script is going to result in massive middle class job loss. That should be modeled.
One mistake in the model was assuming "basic job" would produce anything at all. I suspect the most likely outcome of putting one dude in charge of two dudes, one digging holes, and the other filling the holes in, is both guys are going to pay a small bribe to the boss to falsify the paperwork. Another severe attitude problem issue is you assign a guy to a "basic job" picking up trash in a park, well, if he goes fishing instead, you can't fire him from a guaranteed job. You're supposed to be a teachers aide at school; well, yes, but I'm checking facebook on my phone instead, what are you going to do about it? I suspect the value produced will approach zero.
Another model error is assuming the only unusual payoff will be JKRowling style, but I suspect you'll have 100000 amateur theater geeks and wanna be musicians for every billion dollar author. So the cultural production should explode. Hard to say if it'll be worth anything, but they'll be a lot of it, that's for sure.
I think you and I disagree on this assumption. For one, we still have homeless people, although for a handful it is a lifestyle choice. I'd say it's far more nuanced and complex than to just say everyone can already afford basic goods. If someone loses all their current gov't benefits, but starts receiving $1,000/mo which they are free to spend however they want, how do their buying habits change? No one really knows.
As far as I know, food stamp programs and other current welfare programs have many restrictions on the types of goods you can purchase.
In addition, we aren't considering the marginal costs of increased production of some goods, while others are negligible - e.g., software.
I agree with your other criticisms of the model though.
How many people start new companies since they have a little fallback support? (potential increase or decrease)
How many dual-income families can afford to have a single parent working with the income of two families coming in? (potential decrease)
How many people can afford not to work while in college now? (potential decrease)
In previous question, Does this mean their long term chances of success are higher? (potential increase)
How many people could use the benefits after a layoff + savings to better them selves and their future before re-entering the work force? (potential increase)
This is partially modelled by the jk_rowling() function. If you feel that is insufficient (I agree, it probably is), add a term you feel is realistic and check if it changes the result. For any parameters you don't know, choose a probability distribution that you feel encompasses all plausible values of the parameter.
Then run your new model. Who knows, it might be the case that no matter what plausible value you choose, it doesn't matter. Or it could change things entirely. These are both useful things to find out, and we'll only know if you write some code.
I'm sorry you felt I needed a downvote for offering suggestions for additional variables (I do not think the jk_rowling function alone solves those issues) was a negative comment. I was trying to help without having the skills to directly participate.
Nevermind the dual Masters in Business or the Minor in Economics. Oh and ignore the fact that I talk to my brother about complex economic issues (he has a Masters in Econometrics).
Ignore the fact, I think this is fascinating topic and have tried to read everything I can on the topic.
I'm sorry I tried to help without code.