= 27d567bc2d48d9f40e9ccc20ea370b15)
eli_sennesh comments on Things I Wish They'd Taught Me When I Was Younger: Why Money Is Awesome - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (234)
I've got a couple things to say here.
Most of us really are income satisficers, not income optimizers. That is, if you offered me a $350k/year salary to do my current job (grad-student), I would end up distributing most of that money to charity. Not that charity isn't great, but you could have just given the money to charity yourself instead of taking the wasteful intermediate step of paying me.
Why is this true? Because we train ourselves to make it true: frugality is a large component of financial responsibility. For instance, I'm so used to living at a grad-student's standard of living that money above and beyond the cost of that standard of living is pure luxury: I spend it on frivolous bullshit or put it in my retirement account. I'm fairly sure I'm the only 24-year-old grad-student with a Roth IRA topped-up for 2012-2014, and this is because when I got extra money, it went into the retirement account rather than towards consumption, or even towards extra charity.
HOWEVER (yes, the capital letters are justified there), there is an IMMENSE component of class privilege in this whole subject. I am literally the only person I know with this much money at my age who isn't working for a tech company! Other grad-students have far less money, and everyone who didn't major in Computer Science from my undergrad is semi-impoverished and has student loans to pay off.
And then we talk about the people who came from such systematic disadvantage that they never even got to the point of a university education in a potentially lucrative upper-middle class field in the first place. There are millions of such people, far more than there are of us computery types, even if we restrict ourselves to only consider First World countries. Income inequality is high, in some places higher than it ever has been, and rising all across the developed capitalist world. Public support for things like medicine, housing, and education is falling, with the result that people are finding themselves having to dip deeper and deeper into their private funds just to make it through life.
If we want to talk about how money is awesome, we should also be talking about how to make sure that people who aren't like us actually get some of it. I mean it: no amount of Tumblry "checking our privilege" is actually going to make the people who work in, say, our favorite food trucks or hole-in-the-wall falafel joints, anywhere near as wealthy as us. We'll need to actually exercise our intelligence for that.
This is true for me as well (I'm slightly older), but I also have some sources of income that I expect most graduate students don't.
But one of the main reasons why money is awesome is because spending money is rivalrous. My primary expensive hobby is art collecting. I have the number of original paintings I have because I put up more money than the other people bidding on them, and if everyone had more money, then the primary effect would be that the prices increase.
When you say we need to exercise our intelligence, let me talk about Franklin Barbecue in Austin. It's quite possibly the best barbecue in the US, and they've sold out of brisket every day that they've been open. Officially, it opens at 11 AM, but generally people recommend that you show up at ~8 AM to wait in line.
To the economist in me, this is a terrible setup. They could spend their customers' extra money; they can't spend their customers' wasted time. They should auction off the barbecue, which will raise prices and lower wait times. But it'll also get rid of the communal experience of waiting in line, and less of their customers will be students and more of them will be engineers. The way to get more money to 'food trucks' is to embrace the inequality that makes engineers that will bid on barbecue.
The marketer in me suggests you're off the mark. How do you know that Franklin's bbq is the best in Austin? Because there is always a line and it sells out. The wait in line IS what differentiates their product, and its how people judge the quality in such a subjective market.
I imagine if you start an auction for the bbq, what you'll find is that in a few years you are making less money, as instead of being a good bbq experience that people drive in from all over Texas to try and tourists flock to, you'll be just another good bbq place in Austin.
Them are fightin' words, y'know... :-D
It's more complicated than it looks
I'm aware, hence the hedging. I am not a food critic, and am relying on the judgments of food critics.
Yes, people rage at high prices, especially when demand jumps and supply falls. And I'm sure that the status threat of the price rising or being priced out makes it worse than just the scarcity.
But the right answer probably isn't lotteries. People are unhappier when others receive rewards for merit than they are when others receive rewards because of luck. The right answer almost certainly is efficiency.
This confuses me. Surely if people are made less unhappy by a luck-based distribution, that's an argument in favor of a luck-based distribution?
I'm not sure if you typed that backwards or not. I can think of plausible reasons for people to hate both luck and merit distributions.
I view it as an argument against the preferences of people.
So you did mean it as written. I'd kind of like to see the studies, if you have a link. I don't find it surprising, exactly, but it's not a question I'd considered before, and it seems like it would be amusing misanthropy fuel.
Maybe people don't actually believe in merit, in near mode. Maybe they think they do, but they are really thinking about status.
Distributions based on merit (that we don't recognize instinctively) simply seem unfair. Distributions based on tranparent luck seem like everyone at least had a fair chance.
Maybe the real problem with money is that it usually belongs to people we personally don't know, so we don't know what exactly they did and why exactly should we respect them, so it feels like they really don't deserve the money. And the rest is rationalization.
This is made worse by money anti-correlating with status when all other variables are controlled for, i.e., given two otherwise comparable jobs, the lower status one will pay more.
The right answer to which problem exactly? Temporary shortages of high-status goods aren't exactly a burning issue that really needs to be solved externally.
Locally, the Barbecue Distribution Problem. Globally, the Efficiency Problem. Imagine Franklin Barbecue as one of the broken windows of inefficiency; yes, it only wastes tens of years per year, and they're probably only losing tens or hundreds of thousands of dollars in revenue per year. But efficient markets in barbecue help make efficient markets in other things more reasonable.
I would assume that people who run the barbecue are (1) Aware of the problem; (2) Have incentives to deal with it; and (3) Are not entirely stupid. Given this I am not sure why do you think that what they are doing now is not "the right answer". For example, raising prices might be good in the short term but turn out to be a very bad idea in the medium term.
What is that problem and, again, what does it have to do with temporary shortages of high-status goods? And I'm less than convinced that the broken-windows theory applies to global efficiency. In any case, if so, wouldn't you want to start with government, instead? X-/
To give a trivial example, creating such a temporary shortage is popular marketing trick (if the company can pull it off, of course).
I think that (3) is not a good assumption to make, and I wouldn't word it that way. I know lots of artists who have never heard of sealed second-bid auctions (also known as Vickrey auctions), despite those auctions being the optimal way to sell artwork or commission slots online. Are they entirely stupid? No; they just have limited knowledge. Similarly, the barbecue auction problem has a potentially nontrivial complication: there are 5 different varieties of meat sold by the pound (and each variety of meat can either go into by-the-pound orders or sandwich or plate orders), and many people would like either their entire order, or none of their order. How do you find the optimal set of orders to fulfill, and what price do you charge people for those orders, in a way that doesn't skew their bidding incentives?
It's a solvable problem, of course, but it's the sort of problem you'd want to hand off to an optimization guy to solve for you, especially if your core competency is barbecuing meat.
It might- it's possible that once people could get it by paying more money, instead of more time, it would lose some of the specialness and people would go there less. But it's not clear to me that they would ever reach the point where they don't sell out of meat, and maybe they have to be open for dinner too instead of just lunch.
But it could also be that the steady-state long-term price of their brisket is $40 a pound, and they've been selling it at $17, and that it is a fantastic thing over the medium term.
(Also, I feel I should mention, since it may not have been obvious: they do allow pre-orders, if you're willing to pre-order by about a month. The amount of pre-orders they allow is obviously capped, so that there's still BBQ available day-of. Auctioning off meat should start as a small percentage of their total quantity moved as a test, and then expanded or contracted as desired. So long as some of it is available by waiting, it is unlikely to lose the popularity.)
Basically, not enough people thinking like economists.
Well, then, I see an excellent opportunity for you. You mentioned that they might be "losing tens or hundreds of thousands of dollars in revenue per year" -- surely if you go talk to them and point it out, they'll be glad to pay you some of that surplus that they are leaving on the table.
In the best case you'll earn a fair chunk of money and make friends in the BBQ business. In the worst case you'll learn a valuable lesson why theoretical economics doesn't apply to real life too well :-)
That's complicated. I understand what you are trying to say, but "thinking like an economist" is not an unalloyed good. For example, consider that economics (especially macro) is really bad at forecasting.
This is, in fact, my plan.
thanks for the link. I wonder if it would be feasible for companies to run lotteries for highly demanded goods.
In our economy most shortages are temporary -- the market takes care of them.
But lotteries for things in very limited supply certainly exist, see e.g. this
It's not exactly bidding, but I happened to read an article this morning about the $4 artisan toast available in a certain San Francisco coffeeshop, which seems to largely fit the bill. Yet about half the article was taken up by tedious kvetching about how the Bay Area tech industry is driving up the standard of living.
One could argue (and indeed I largely agree) that this ignores the artisan toastmaster side of that economic equation, or raise any number of other objections, but that's not the point. The point is that this is a tough political sell.
(Statement of conflicting interest: I am in fact an exploitative Bay Area techie.)
The Marxist in me wishes to point out that if you're part of the tech salariat rather than the VC class or real-estate rentier class, you are not in fact exploitative. The fact that the Bay Area confuses "high productivity worker" with "exploitative capitalist" is one of its larger collective errors of thinking.
In the wage / rent / interest model, the skilled person's salary should probably be modelled as a mix of wage and rent.
The wage in its pure form is what a completely replaceable employee gets. The fact that the employee is completely replaceable will drive the wage down to the level where it barely covers the expenses to survive. Of course the expenses are different at different places, so the wages will reflect that, but that additional money just goes through you, and at the end you don't benefit from it.
The rent in its pure form is what you get for auctioning a use of a scarce resource (such as land). An intelligent person with mathematical skills good enough to work in IT is in some sense a scarce resource. They can be replaced (but you can also move from a piece of land to another piece of land), but it's difficult, and there are not enough skilled people for every employer's every whim. The employers are competing among themselves, and this creates the rent. -- If you could somehow separate your talent from your person, and send the talent to the work while you stay at home and have fun, that would be a rent in its pure form. But because it doesn't work this way, your wage and your rent are connected together.
And the interest in its pure form is money making another money. Which you can achieve by investing your rent in an index fund. So a techie can become an evil capitalist, too; it just doesn't happen automatically and requires some strategic thinking.
Ah, that was a pleasant bit of reminder. Thank you.
(Though I did mention I have a maxed-out Roth IRA. I very much am trying to become an evil capitalist, on grounds that within this system it is my only rational move, should I desire to do anything other than work for a minimum subsistence. I still want the system changed and overthrown.)
The toast is Josey Baker Bread (yes, that's actually his name; short documentary here) and it really is that good. By which I mean, as another exploitative Bay Area techie, I've paid that price at The Mill more than once and I felt it was worth it.
The toast only manages to be worth four dollars to you because four dollars is worth less to you than to poorer people. (At least in the sense of how much you would care if you lost four dollars and what you would be willing to do to get another four dollars).
Some customers don't have extra money, but do have extra time. (And they can't easily just convert the extra time into money.)
Yes, but when you have more people who want to be customers than who are customers (as is implied by selling out of stock before you run out of line), the question is "who do you exclude?"
One way is exclude people randomly. This gives everyone involved a sense of fairness, but no sense of control- either they get lucky, or they don't. Another way is to exclude people by time preference. If they aren't willing to wait, then they aren't going to get any. A third way is to exclude people by cash preference. If they aren't willing to pay the price, then they aren't going to get any.
Generally, the only option that benefits the supplier is the last one, and thus it's probably the one that they want to take. There are a few counterexamples; one example that frequently comes up here is concert tickets for young female singers. There are generally two kinds of fans that go to those concerts: young girls who enjoy the music and older men who enjoy the show. The older men are generally willing and able to pay more, but not as willing and able to wait in line. It's unlikely that the optimal experience for all involved is for all of the tickets to go to the older men, and so they might parcel out tickets to different venues in the hopes that they will go to different people while still increasing revenue.
If they raise the price of the barbecue until the number of buyers is small enough to eliminate the line, their pricing will be seen as unfair and this will have a long term affect on their ability to retain customers. While barbecue could produce more profit being sold to rich customers, other more plentiful items need to be sold to both poor and rich customers to maximize profit. And if you lock poor customers out of the barbecue, they won't come in for the other items.
Furthermore, it would be rational for poor customers not to buy the other items because of the transaction costs in having to discover which items are priced for rich customers and which aren't. The store could do slightly better by pricing all items for rich customers rather than just the barbecue, but even that may produce less profit than keeping the price of the barbecue low and having more customers for the other items in the store.
Another possibility is that people get tired of eating barbecue too often and you need to have customer turnover. If you price the barbecue high enough that exactly the number of rich customers arrive that will buy all the barbecue, in a week from now those customers will be tired of barbecue and there won't be other customers to replace them.
Sure. But consider airlines, and the revenue management they do, as a contrast.
The basic problem is that there is no single price-per-seat at which it is profitable to fly a plane. Imagine the demand curve as something like $1000/x, where x is the number of tickets sold on the plane. Regardless of the price you pick, your total revenue is going to be $1000, and if the plane costs $2000 to fly, you can't pick a single price for every ticket such that the plane is profitable to fly.
But suppose you could offer different customers different prices. The person willing to pay $1000 is charged $1000; the person willing to pay $500 is charged $500, the person willing to pay $333 is charged $333, and the person willing to pay $250 is charged $250. Now you've got a plane in the air, and $83 in profit (and another person paying $200 would get you up to $283 in profit). But this required you knowing which customer was willing to pay what, which is generally done by time-segregation (the amount of time you book the flight in advance, combined with the number of seats left on the plane) which is itself determined by sophisticated modeling.
There doesn't seem to be a public outcry about revenue management for airline tickets. Perhaps this is just because people don't understand what's going on underneath, or adjusting prices with time feels appropriate in a way that adjusting prices with supply doesn't, or because it's been this way for ~30 years and people are used to it now.
This is not a problem for auctions, because the price drops when the demand drops, so long as the minimum price is set so that the market always clears.
Analogically with the airlines, the current model should be the "economy class" barbecue, and there should be a new "business class" barbecue -- extremely expensive, but without having to wait.
Preferably with some additional differences -- sitting in a separate room, with pleasant music and paintings on the wall -- to make it easy to rationalize (by both kinds of customers) it as "paying extra money for extra luxury" instead of "paying extra money for cutting in line".
That model is used by Disneyworld and other theme parks. You can buy a regular ticket, or you can buy a premium pass which costs more but gives you the right to skip the lines at the attractions.
This isn't actually the case at the Disney park in California (not familiar with anywhere else). There are different season passes, but the premium ones just let you get in on weekends and holiday days and what not.
They do have "fast passes" but those are available to anyone- you go to a kiosk and get an appointment to come back to the fast-pass line at some later time.
Legoland, for example, sells a Premium Play Pass which gives you "front-line benefits". Universal sells the Express Pass which allows you to "skip the regular lines".
Disney, I think, is more wary of PR problems, but still you can buy the (very expensive) "VIP tour" which, as I understand, will allow you to ignore all lines.
Customers prefer constant prices. Aside from the perceived unfairness, there are, again, transaction costs. Any time spent by customers trying to figure out how to get the lowest price is still a loss.
Airlines get away with it because airline seats are in limited supply, making it a seller's market. The buyers have to take whatever the airlines give them. There's certainly a fair degree of public outcry about it; the fact that there isn't more is because of a combination of people not understanding it, the fact that most people only buy airline tickets occasionally, and the fact that there's nothing the public can do about it.
Customers prefer predictable prices. They don't have to be constant.
E.g., if tickets for an event are $15 in advance and $25 at the door, and this is stated clearly up front, most customers are OK with that... we can plan early and save $10, or we can keep our options open and pay a premium for that privilege.
On net, airlines lose money. In recent years, it seems to be mostly because of decreased demand (due to terrorism fears and TSA harrassment of passengers), and for decades it's been because of price wars between airlines. I don't think this is a market best described as a "seller's market."
That's a very... incomplete prior. Customers also prefer cheap prices. Customers prefer (a lot!) the feeling that they got a deal and bought something on sale.
I also don't see what's special about airlines. Pretty much every business would love to price discriminate. Many do through a variety of methods. For example, supermarket coupons are a classic form of price discrimination.
This assumes that a) there is a fixed supply of original paintings, and b) the demand for original painings is income inelastic. Admittedly, I'm not an expert on the art market, but my intuition is that the opposite is the case on both counts: as incomes rise, I would expect people to spend a larger percentage of their income on luxary goods such as art. If this is the case, then, yes, everyone having more money would indeed cause the price of original paintings to go up, but they would rise at a faster rate than less elastic goods, which would cause production of said paintings to go up, which would drive prices back down; the net effect is that more people have more paintings.
I decided to not elaborate on that because the second-order effects depend on why everyone has more money. If it's because everyone is more productive, then there's also lots more art floating around, because the artists are also more productive. I do agree that people who are richer spend more money on luxuries like art, but it's not clear to me that all ways of giving people more money actually make more rich people.
But even if there's a bunch more art floating around, there is a fixed supply of the best original paintings, and those will still go to whoever wants to spend the most money at art auctions. (Of course, best is subjective, and so on, but that's part of the point of using auctions.)
"Some" of it they generally do. But if you're going for equality, do note that people are usually paid for the value they produce and that individuals' capability to produce value differs GREATLY. Even if you control for things like socio-economic status.
I half-agree. I'm actually starting to believe that factors like trade, industrial policy, and public regulation of economic rents and public goods (take the preceding concepts apolitically, for the moment, please) have more to do with our current economic crises than any notion of individual "merit". That's not to say there's no such thing, merely that in particular, policies regarding trade, industry, and economic rent seem like much stiffer variables than the relatively loose factors of individual work-ethic or education, or even things like national work-hours.
For instance, a country that exports large amounts of capital-intensive goods while strongly regulating its financial sector (say, current day Australia or Germany) seems to be able to afford uneducated individuals, expensive social programs, or short work hours much more easily than a country that theoretically has higher per-hour productivity but suffers a trade deficit and has largely financialized its economy (say, current day America or the UK).
What we end up with is that America and the UK suffer massive income inequality, while Australia and Germany are more equal and stable -- even though they're all First World countries with their own top-level educational institutions, labor expertise, and companies. A theory which treats macroeconomic policy as a stiffer (more strongly predictive) variable than individual/company-level merit therefore seems more likely.
I don't understand what do you mean -- I can't see any connection between "individual merit" (and by "merit" do you mean the productive value of a person?) and current economic crises.
I don't understand that either. It's not that, say, Germany can afford a more generous welfare system than the US -- after all per-capita GDP is higher in US than in Germany -- it's just that Germany chooses to reallocate more of the wealth produced in this way.
Equality isn't a good yardstick -- the old USSR had much more equality than any Western country. And I don't see the stability you're talking about. Stable in which sense?
Some might say that exactly that is the problem. (Not necessarily myself.)
To these people I would point out the difference between reality and various imaginary universes one can construct.
Is that a new version of the is-ought fallacy?
Kinda. It's really more of a confusion between what the world is and what you would like it to be.
do note that people are usually paid for the value they produce and that individuals' capability to produce value differs GREATLY
That's true, but I still care about people who don't produce much value, and I don't like to see them being impoverished and miserable.
Sure. Nobody says you have to not care about less productive people.
So redistribute some of your value to them.
The obvious answer is: Make as much money as you can, retire early, and then spend your time solving this problem.
For example, you could start giving poor people free programming lesssons. Or even donate them computers. Or think about something smarter and more effective than this.
I've seen this suggestion elsewhere. I'm all in favor of it, but it kind of bugs me anyway. The assumption is that most people can learn to program (or do other forms of IT) if taught. I don't think that's the case. Programming well is hard. IT pays reasonably well because good IT people are hard to come by, and I don't think lack of access to training or facilities is the reason. Certainly not since OSS became widespread.
Any widespread solution to poverty has to work for people that don't have hacker-natures or Mensa-class IQs.
Yeah, it's a solution of type "pick a few people who are easiest to save, and save them, ignoring the rest".
It's worse than solving the problem globally; and it's better than doing nothing -- which is what most people will do; including most of those who like to think about global solutions. It's far from optimal. It's also something that one can do as an individual -- so it can be used as a backup plan is case no better idea comes around.
Assume that 1% of people could become good programmers. If we trained (or offered training to) 10x as many people, we would still end up with 10x as many programmers.
I grew up with computers in my home; I had a programmable calculator in middle school; my high school offered programming courses; my family could pay for me to go to a very strong CS university. Not everyone has those opportunities.
That assumes a random distribution of potential programmers. Isn't IQ highly correlated with both familial wealth and programming ability? I doubt anyone's compared the programmer-nature with wealth directly, but if the two are also highly correlated, that 1% could mostly already have access to the training they need. Offering to 10x more would just get you (slightly less than) 10x more bad programmers.
Or make as much money as you can and then spend your money solving this problem.
I actually tried to donate to them, and they claimed my card was declined. So does NewEgg. I think some of these processors aren't equipped to handle debit cards, apparently, since both Amazon.com and CareerVillage have proven capable of "shutting up and taking my money".
Further pity, the Rolling Jubilee had stopped taking funds for the year. They're a favorite of mine: they buy up defaulted, second-hand debts of the housing and medical kinds that cripple people so damn hard, and then just abolish them. People who were being hounded by collectors get a call and find out their catastrophic debts are just gone.
Don't they then get a call from the IRS saying they have to pay taxes on this gift?
That sounds like it might be a frequently asked question! Let's check Rolling Jubilee's FAQ... yup!
Point being, if you can think of a difficulty in the solution of a problem in two seconds, then the people actually working to solve the problem have almost certainly either anticipated or run into the difficulty and have dealt with it.
Point being, it's not up to the Rolling Jubilee to decide. It's up to the IRS to decide.
If the Rolling Jubilee wants to be serious about it, it needs to get a Determination Letter from the IRS stating that yes, IRS will treat these debt cancellations as tax-free gifts.
Yup. Have a look at that interview with the top tax lawyer to see how that will likely play out. (Short version: 1. Medical debt is likely much easier to forgive tax-free than mortgage debt. 2. The debt forgiveness is currently so small that it will probably be lost in the shuffle without the IRS wanting to devote resources to making a determination.)
The point of my comment was really that if James_Miller had posed the question out of genuine curiosity instead of as a rhetorical side-swipe, he could have found the answer.
Fantastic! Now I can escape the tax liability on my paycheck by having my employer pay my debts rather than directly paying me a salary. Or, rather than contribute to MIRI and have some of this money go to salaries which are taxed I could instead payoff some of the debt of MIRI employees.
Sorry for the sarcasm, but for someone who has studied a bit of U.S. tax law it's obvious that the IRS can't allow Person A to payoff the debt of Person B without there being any tax consequences. Doing so would just create too massive a loophole for tax avoidance.
I think Rolling Jubilee has a better case for forgiving debt as an "exempt purpose" than you're giving them credit for. In any event, EY noted recently (on Facebook, possibly?) that the legal system is made of people, not words. An obvious cheat won't cut it.
(I can hardly complain about the sarcasm, having dished some snark of my own. It's all good.)
It seems that (1) in practice, so far they apparently don't, but (2) no one is quite sure whether they should. (Or, rather, a few people are sure they shouldn't, a few more are sure they should, and lots of people aren't sure.)
A potentially easy way around the problem is to donate through GiveWell. Failing that, you could try via Venmo.
Damn!
Bloody hell!
http://i.imgur.com/ByU9iq0.jpg
Perhaps you should get an actual credit card as well as your debit card, for use with vendors and charities that can't handle debit cards.
(I appreciate that there are a number of possible reasons why you might be unwilling or unable to do that. They may be very good reasons. It's probably worth spending a few minutes, if you haven't already, considering whether they are good enough to outweigh being unable to donate to entities you would like to donate to, etc.)
I had no problems donating to GiveDirectly with my debit card; OTOH now I'm having trouble giving to CFAR via PayPal using the same card, which had never happened before to me. (I guess I exceeded my daily quota for Internet purchases.)
The thing is that it's supposed to function as a Check Card, able to be swiped as both credit or debit. This is actually the first time I've ever found anyone who won't take it.