Economists Discover the Poor Behave Differently From the Rich
businessweek.com
businessweek.com
Apropos of the article: the headline is really making hay out of a tangential point. Sure, the "poor vs. rich" angle gets more clicks and inbound links. But the real point -- which is a huge one, as far as the field of economics is concerned -- is that the "rational (and by implication, homogeneously rational) actor" presumption is under fire. It's been under fire before because people have been shown, time and again, to behave economically irrationally. But this time it's under fire because people have been shown to react heterogeneously to the same stimuli and incentives. (Not really touched upon in the article, but also worth mentioning: a third major critique in recent years has been the revelation that classical economics was highly skewed by the idiosyncrasies of the Western cultures in which it was cultivated and codified. It turns out that not everyone around the world behaves like 19th-20th Century Austrians, Britons, or Americans.)
All of this is huge, I repeat, because pretty much every foundational theory in classical economics is predicated on homogeneity and rationality. Every single one. The entire field. Even a preponderance of modern economics is undergirded by flawed assumptions borrowed from classical models.
I, for one, am really excited by the modernist, behaviorist, relativist, empiricist and frequentist challenges to classical economics. This is sort of like seeing a paradigm shift in a highly influential field of study play out in something resembling real time. Economics has always been based on statistics -- even highly sophisticated statistics -- but the inputs, so to speak, have been pretty unsophisticated. These challenges will help drag economics into the 21st century.
If I were an expert in how bubbles formed in water and spent all of my time thinking of ways to maximize bubble growth by sucking gases out of the water in a tank, I would be good at creating aquariums with large bubbles, but probably end up with a lot dead or sick fish.
Also, "rational" is contextual, not objective or axiomatic in any real-world finite computational system as a matter of mathematics. This is an important distinction often ignored. As a consequence, "irrationality" can be generally defined as "making a decision from a different contextual basis". Again, this defeats the desire to create absolute rules of economics that fit in a soundbite.
Economics is perfectly capable of dealing with the complexities of real-world actors in theory. In practice, most economists attempt to reduce the complex dynamics implied by the theory into trivial theorems that are easy to turn into government policy. This is somewhat understandable because even approximating a pure model is computationally prohibitive. This provides a lovely framework within which any economist can come up with a narrow model that suits their particular ideological proclivities. You rarely see an economist that properly recognizes the constraints and assumptions of their model but those limitations exist nonetheless. Advanced computer science will allow us to implement the more general models directly regardless of whether or not the results are politically convenient, but it may take a while before we see it.
Likewise, choices that appear irrational may turn out to be entirely sensible but for the exclusion of relevant factors. Having been quite poor in the past I can certainly distinguish between irrational choices of the 'oh fuck it, let tomorrow take care of itself' variety and rational ones of the 'this is inefficient, but satisfies constraints' type.
As an example of the latter, consider riding the bus. In many places you can get a monthly pass at a substantial discount to paying multiple fares every day. But for someone who is un- or under-employed, the expected value of purchasing the monthly pass is quite different from that of someone with a predictable commute; if that person makes the wrong guess about how often s/he will travel over the month, it can result in a significant opportunity cost.
Then there's the even simpler case of the person who can't accumulate enough cash at a time to buy the monthly pass, but who won't be able to get paid at the end of the week without taking the bus to work each day. Absent credit facilities, paying bus far on a day-by-day basis can be perfectly rational when viewed in the individual's overall economic context. Unfortunately, economics textbooks don't devote much attention to multiple inelastic constraints.
Having a high discount rate is not irrational. Multiple discount rates are, but a single high discount rate is perfectly rational.
But if you discount the future too liberally, then pretty much anything can be spun as a rational choice, eg drug addiction, being a hobo, taking up crime or whatever. I smoked fairly heavily for about 20 years, which I'm sur is going to have some negative impact on my future lifespan/quality of life, if the way I felt by the time I quit is anything to go by. For all that I liked smoking at times, it was basically a symptom of nictotine addiction and I can't say my life was improved by having been a smoker. This is certainly one case where I discounted the future too heavily.
Not everything we do is rational, but all the things you mentioned can -- and imo should, absent evidence to the contrary -- be treated as rational choice.
Er, no, while people have been saying that this is the "real point" of various stories on internet discussion fora at least as long as I've been on internet discussion fora (which goes back to the latter half of the year 1990), its really not. The rational actor model (whose role in economics is much more limited than many people seem to think) has always -- from the day it was conceived -- been known to be a simplified model (like the "perfectly competitive market") which is a good-enough approximation for some real world interactions but fundamentally not true (it assumes, among other thing, what amounts to effective omniscience -- at least where it concerns any decision that has any impact on the actor's own experienced utility -- on the part of all actors), and also not even good enough to be useful for many important situations.
No one anywhere in the social sciences thinks the rational actor model represents the underlying reality, and in economics particularly it (along with perfectly competitive markets) is baseline that works both as a basis for a highly idealized model that is important pedagogically and often useful as a basis for describing real world situations in terms of variances from (e.g., explaining empirically derived models of real-world behavior in terms of how the empirical model differs from idealized model and connecting that to a theory of how the real underlying processes differ from those that form the basis for the idealized model.)
Part of this may be due to the fact that most economists who appear in the media are basically paid shills, but no matter the reason, it is definitely a problem for the democratic process.
Preference heterogeneity by itself does not imply anything about rationality. However, that is not to say the topic is so simple. We know this empirically by looking things like revealed preference for risk taking. One example might be propensity to take risky bets, despite known zero/negative NPV expected values. These people are happy to buy earnings Vol because the distributions on the downside has such little variation (poor today = poor tomorrow). Whereas a middle class person can make more use of the savings (investment opportunities are greater, and the downside risk of less savings is too), on the one hand, and is not so desperate for upside on the other (even if its just the psychic hit of the "chance" to be rich).
I have absolutely no idea where you get this idea. Both supply and demand curves are based entirely on the concept of heterogeneity.
The microfoundations of the supply and demand curves are (for simplicity assuming each customer will supply/purchase only 0 or 1 item):
Supply(P) = # of suppliers with reserve price > P
Demand(P) = # of customers who's reserve price < P
Similarly, the standard economic explanation of why the poor make "bad" choices is heterogeneity - the poor have a higher discount rate than the rich.
The question is, how much, and what kind, of irrationality is reasonable to admit into a model? While rationality is easy to define, there can be an infinite number of ways people can be irrational, and their decision processes would change. Macroeconomics used to assume less-than-rational, empirically calibrated decision processes and aggregate behavior that would rise from those; but these models did not adequately account for people changing their behavior, including the degree of irrationality, in response to changes in the environment.
First, economics as a field has somewhat fallen into a trap where people are overly focused on models that are analytically tractable. This leads them to use unrealistic models that can be solved with infinite horizons, rather than realistic models with finite horizons.
Second, there is always a limit to how realistic your model can be, so unless there is a strong reason to think that heterogeneity is important in your model, you probably won't include it. In this case it may not have been completely obvious that heterogeneity was really important in analyzing the effects of stimulus.
Finally, most people with economics degrees are about as privileged as people with the same degree level in other fields. So being privileged really cannot explain disagreements between economists and other educated people.
But these "wrong" facts are useful when creating predictive models of our universe. That's all this is - a "wrong" fact used to help simplify a model. Turns out the representative agent might be an oversimplification, but so are G (9.81 m/s^2) and c (299,792,458 m/s), depending on what you're doing.
In a "science" (science is a methodology and lies a spectrum, and is not a dogma, so let's try to keep some perspective) where you don't really have a "lab", all you can do is create predictive models that fit your observational data.
Technology of the past 10 years brings more data, and lets economics move away from the representative agent of the past 100 years. Good. Let's do that, but let's not forget whose shoulders we sit on - no one mocks the orbital model of the atom, or Newtonian laws of physics, even though both models are outdated [2][3].
[0] - http://curious.astro.cornell.edu/question.php?number=465
[1] - http://arxiv.org/abs/astro-ph/0305457
[2] - http://spiff.rit.edu/classes/phys314/lectures/bohrprob/bohrprob.html
[3] - http://en.wikipedia.org/wiki/Classical_mechanics#Limits_of_validityIt is astounding to think that changing your personal spending based on what happens in DC is supposed to be the "rational" behavior. These are not economists, they are radical political ideologues dressed in pretend academic clothing.
The individual doesn't have to deduce this themself. The information could come second or third hand through a variety of channels.
Just a few years ago data from lab experiments was not widely accepted because people were unsure if it was relevant to how people act in the real world. Countless times the effectiveness of lab experiments has been proven.
I am looking forward to the day when people finally accept that traditional economic models are not perfect (duh!) and that people are not perfect, rational, decision making robots.
There are huge cultural differences also. Russians are spending enormous amount of money on their mistresses using easy bank credit they barely able to pay back, while Cambodians are buying Lexus cars while living in wooden shacks, etc.
While so-called poor people (read proles - ignorant working class with very restricted, repetitive behavior) are much more prone to the peer and the bandwagon effects and much more concerned about "equality" and conformity, no oversimplified economic model could capture a complex and volatile behavior even of a primitive drunk.
My guess is that so-called "behavioral economics" and "behavior finance" are attempts to make more realistic and useful models, together with an important notion that the whole system is extremely dynamic, volatile and constantly changing/evolving. Equilibriums are moving.
"The challenges the world faced in 2010—low home values, credit hard to get, central banks unable or unlikely to lower interest rates further—“were nonexistent in representative-agent models,” says Christopher Carroll, who teaches at Johns Hopkins. At the ECB’s Frankfurt conference, Carroll presented a paper that bolstered Krusell and Smith’s model with microeconomic data. For Carroll, people differ in one crucial way. “The marginal propensity to consume,” according to the paper, “is substantially larger for low-wealth than for high-wealth households.” Rich people behave like the hyperrational agent. They plan for the future. They save during a stimulus, thinking about the taxes to come. And they can borrow during a fiscal contraction.
Poor people are what economists call “borrowing constrained.” They tend to have more needs than are being met, so when money arrives, they spend it. When the government stops spending and credit is hard to come by, the mythical everyman, like the rich person, continues to spend. But most real people don’t have access to credit, and they hunker down. Carroll’s findings have been confirmed by other academics in the last two years who looked at Italian and U.S. data."
I remember what poor was - I always knew where my paycheck went well before I got it.
1) they know, but it isn't easy to demonstrate, so it doesn't get published and therefore it doesn't exist because nobody can provide a citation. Alternatively,
2) most people who have economics degrees are fortunate enough never to have struggled with serious poverty
A major issue is that detailed economic models are verified against the limited time period where we have detailed data, so if some factor X is constant during that time, then you can't really know if its modeled appropriately.
One problem here is that actually, heterogeneity does not really matter much for the aggregate! In other words, the model may be more realistic, but the aggregate behavior -- such as consumption and economic growth -- tends to be relatively unaffected by inequality. This happens because of two things: (1) the poor people, exactly because they are poor, do not consume enough to create a meaningful economy-wide impact, and (2) people really hate being poor, and save more during good times so that they don't get hit with the constraints in bad times. The second one can be to some extent accounted for by allowing a sizable fraction of people to be somewhat irrational, and by getting rid of the "infinitely lived" consumers; the first would stay though. I am curious about what is exactly done in the Carroll paper to make it matter.
The IMF is a criminal organization.
Never ascribe to malice what can be adequately explained by stupidity.
Do you seriously think you can just copy paste a phrase that sounded about right in certain context and reuse it everywhere?
"Many journalists have fallen for the conspiracy theory of government. I do assure you that they would produce more accurate work if they adhered to the cock-up theory."
How's that holding nowadays? What a useful phrase!
Besides, there's also this variation:
"Never attribute to malice that which can be adequately explained by stupidity, but don't rule out malice."
But it doesn't sound as smart to little kids, and it doesn't simplify your life by letting you skip all the thinking, right?
Or to put it another way: The road to Hell is paved with good intentions.
For example, suppose there's some 20% off offer available on a regular household expense. The rational thing to do is buy in bulk so you can take advantage of the offer, but if you're poor that option isn't necessarily available, because buying in the requisite quantity to receive the discount means going without some other necessity - eg there's no point in buying lots of diapers when they're on sale if it means there's no money left over for baby food.* Of course couponing allows for fairly flexible strategies and with time the savings can add up, allowing for some greater flexibility; but the underlying fact is that the more money you have, the easier it is to take advantage of discounts that save you money. Inability to access economies of scale is a brutal fact of life at the bottom of the economic pyramid that manifest almost every time one needs to buy food.
* It's just an example, don't get hung up on what exact level of public assistance is available to people with babies in an attempt to negate the point.
For instance, it is very difficult to reduce debt, invest, obtain an education, or otherwise "get ahead". All income is instead spent on day-to-day subsistence, so they continue to receive a lower share of the economy, while paying down little principal and paying maximum interest.
It's another overlooked element: it's not just about income and wealth. It's about cash flow. It's just somehow odd that if a person earned the same income over his lifetime, but received a sizeable advance on some portion, it could dramatically alter the quality of his life.
Leave aside experiments on humans and animals, these guys are basically experimenting with families and entire nations just for fun of gathering more "data" for theories they can confirm or deny by just asking people in the streets.
Kind of makes me help understand why Indian politicians were such staunchly opposed to recommendations of IMF and their group of economists all along. Those rich idiots don't know how 99% of the world functions.
http://m.theatlanticcities.com/jobs-and-economy/2013/08/how-...
Why are economists always so hung up on the concept of saving vs. spending, as though one is inherently better than the other?
Spending is good because it stimulates the economy -- if 1000 instances of Joe buy some new clothes then the clothing retailers can hire some new sales staff. Investing is good because it stimulates the economy -- if 1000 instances of Joe buy a share of stock then the availability of capital to create new small businesses increases and new small businesses create jobs.
It isn't about spending vs. saving, it's about whether the money goes to something productive or efficient or it goes to something harmful or inefficient. If a rich person gets more money and invests in small businesses, that's good. If a rich person gets more money and uses it to hoard precious metals, that's bad. If a poor person gets more money and uses it to buy books and go to college, that's good. If a poor person gets more money and uses it to buy cigarettes or to secure credit to buy an inefficient luxury SUV, that's bad.
It's not about saving vs. consumption. Unless someone is just stuffing cash into a mattress (which is bad and also stupid), "saving" is just consumption of investments, which is consumption by proxy when the invested-in entity spends the money. The question is whether the thing ultimately consumed is more beneficial than the alternative -- which is a question I can certainly see why economists like to avoid, because you practically can't answer it in the aggregate.
Here's a great example. The hypothesis is that rich people spend differently than poor people because they're rich, right? The fact that they have money allows them to act rationally because they can afford to make long-term investments. So, if the hypothesis is correct, lottery winners should behave like rich people, because their newfound wealth allows them to plan for the long term and stop making short term decisions just to keep food on the table. Reality? A majority of lottery winners go on to declare bankruptcy.
Because it's not about rich vs. poor. It's about what types of economic activity produce beneficial results. But economists and governments are preposterously bad at predicting what those things will be.
Which means I'm going to be back in here beating the drum on the basic income. It provides the poor with a baseline level of consumption which they can choose of their own free will. The problem with need-based programs is that the government is deciding what people need, and they're terrible at it -- and then they create an enormous marginal tax rate on the poor by withdrawing the benefits from anyone who dares to live the life of luxury that exists just above the poverty line.
Let the people themselves decide what they need. People in the aggregate are smarter than legislators.
Generally, it has nothing to do with normative evaluations of what is better, it has to do with modelling how systems behave, and spending vs. saving are important difference in behavior with substantial effects within the economy (whether the effects of one or the other are better -- whether generally or in some particular circumstances -- than the effects of the other is a separate discussion, the import thing for the science of economics is that the two are significant, and different, and modelling the difference is therefore important to modelling the behavior of the economy in different circumstances.)
Saving decisions by households primarily mean less spending, hence less demand, hence less incentive to invest in the first place. This is a plain common sense first order effect.
That there may be a second order effect via interest rates is somewhat plausible, but it just boggles the mind how many people seem to believe that this second order effect would ever trump the first order effect of savings decisions. And I think this is explained by the simple fact that people do not correctly distinguish between savings decisions and investment decisions, where I mean investment not in the sense of "buy some shares/funds", but in the sense of "create new means of production".
Just like it has been said that physics is understood through the search of the correct sign, it seems that economics is understood through the search of proper definitions of what we mean by "saving" and "investment".
This is just what I'm talking about. Dismissing the economic benefits of increased investment as "second order effects" when what really matters is whether the recipients of the money make more economically productive decisions than would the consumers in the alternative.
Take the obvious example. On the one hand you can invest money in a local small business which would otherwise not have had an investor and will use the money to hire more local employees who will go on to pay local taxes (allowing lower tax rates or more local services) and consume further local goods and services. On the other hand, you can go to Walmart and buy some knickknacks from China, which will do very little for the local economy as the primary result will be to stimulate the economy in China. It seems exceedingly obvious that the investment can be better for the local or national economy than the consumption. On the other hand, you can easily imagine some investment choice that would be less beneficial than some consumption choice -- because what matters is not the distinction between savings or investment and consumption, it is the economic benefit to the local economy of the ultimate use of the money.
> investment not in the sense of "buy some shares/funds", but in the sense of "create new means of production".
That distinction is certainly important, but that's exactly what I'm talking about. If you buy some shares, now the person who sold them has the money. So what does that person do with it? Perhaps they immediately turn around and buy some shares in something else. Now that seller has the money, and so on until somebody does something with it other than buying shares. So what is that? It could be something useful, like some economically productive type of consumption, or increasing the means of production. It could be something stupid, like hoarding gold or sticking it in a mattress or buying a prime piece of real estate and then not using it for anything. But these are the same decisions that the original party faced, and there is no reason to expect that the seller(s) of the shares will be any better or worse at finding a productive use of the money than the original party would have been.
Let's even go ahead and concede that on average savings are less economically productive than spending for some reason. You still have the same problem, because averages are not instances. There remain many instances of saving or investment that are more economically productive than many instances of spending, and preferring the latter to the former is conceding an enormous amount of economic inefficiency.
That is probably a wise decision for that individual, but it's not going to boost the economy.
Your obvious example may be obvious to you, but it is not a typical example of savings decisions. That was my point: You have to distinguish clearly between savings decisions and investment decisions. Also, typical cases and averages are what matters when you look at the macroeconomy, and that's what you were talking about in your first comment to which I replied.
How do you mean there is no recipient? In that case the recipient is the lender or the bank, which allows them to lend the money to someone else.
You might also be surprised how good for the economy paying down debts can be in the long term. The person who paid the debt is not paying interest on it anymore, and has now improved their credit worthiness, which reduces the interest they pay on all of their remaining debts. The interest not paid can be spent or used to pay further debts (virtuous cycle). Or in the future the borrower can leverage the increased credit worthiness into more borrowing at lower monthly payments, allowing increased future spending.
This is really one of those situations where politics and reality collide. Having the middle class pay down their debts is great for the economy in the long-term. But it may not do anything much today, which is what politicians care about when they're running to reelection. Hence we have a bunch of backward-facing policies that prioritize spending over saving when they are both important to economic growth and the choices between one and the other would be more efficiently made by the people.
Similarly, if you save money at your bank, your bank doesn't actually get more money. At best, you're shifting it from a checking account to a savings account.
Because effecting changes in aggregate saving vs spending is one main thing that policymakers use for regulating the economy (to counter overheating or depression).
I would never have guessed that. There's a longstanding line of thought that says rich people are rich because they spend differently than poor people （obviously, I'm simplifying). A finding that rich people and poor people spend differently does nothing to reverse the form of the argument.