« Lump of Labour, Say's Law, and the slope of the AD curve | Main | David Levine's accidental Monetarism »


Feed You can follow this conversation by subscribing to the comment feed for this post.

When I hear the word "social", I reach for my shovel, because I suspect there's going to be a lot of BS.

"I've been asked to write a paper reviewing "social benefits" delivered through the personal income tax system. I have no idea what this means."

I don't either. I would be tempted to just delete the word "social", and talk about "the benefits of the income tax system". It pays for stuff like army, police, roads, schools, etc., which (usually) benefit us. Done.

I would say that (almost) *all* of economics (once we get past the individual choice problem, or Robinson Crusoe) is "social". Economics is about the interaction of 2 or more people. That's what "social" means. But then it's a bit redundant.

It's definitely code for something. "Socialist"?

Great post!

But I think you left out four words: "Our ordinary business of life is interacting with other people as little as possible."

ps, I remember that Far Side cartoon. It was ages ago...

Nick - "I would say that (almost) *all* of economics (once we get past the individual choice problem, or Robinson Crusoe) is "social". Economics is about the interaction of 2 or more people. That's what "social" means. But then it's a bit redundant."

Exactly. Did you notice the subtle shout out ("rules and institutions")?

Arthurian - thank you. Yes, it was ages ago - from the limited amount we know about who tunes in here, readers of this blog skew young (also male and v. highly educated), so realized after posting that probably many readers wouldn't even know what The Far Side was.

I remember sitting in the broken down old couch (it smelled of wet dog and you sank more than sat in the thing) in the ancient Eastern Townships farm house owned by some close family friends reading a Far Side book by the light of the Christmas tree while waiting for Christmas dinner to be served. I must have been 10 or so (so circa 1984). That's more than 30 years ago now (geez, that's depressing). 'Social' was what was happening in the kitchen.

Bloom County was another excellent comic of the same era. One of my favorites has an ecstatic customer at a gas station exclaiming "Dump the milk! The cat drinks unleaded from now on!"

Even though I have quite different political affiliations, I side with Nick. Get the shovel. The Sokal affair went down when I was in University, and it was pretty much the nail in the coffin in my view.

Hmmm. When I read, "The analysis of social practice . . ." I think, it sounds like there are some operational definitions there. :)

Hmm. My inclination is that the request is asking you to consider welfare/transfers from beyond a public finance perspective - i.e., positive effects of the status quo income tax system outside of maximal-efficiency-considerations-holding-welfare/transfers-policy-constraints-constant. I'd look to happiness or social status quantitative research. If you hold the amount of revenue to be raised as constant, there's also comparing personal income taxes to alternative tax regimes, which at prevailing levels of revenue, could be quite damaging indeed.

My second guess is that it's asking you to study programme benefits delivered through tax credits on personal income from the usual marginalist, individualist perspective. Can be tricky, especially when you try to model the welfare impact of programmes with low participation rates. Does non-participation just mean that the eligible household is indifferent, even if the calculated lost benefit is in the hundreds or even thousands per month? Or is the eligibility estimation just dodgy?

" Yes, people make rational choices, maximizing well-being"

Really? Where can I meet these people? Most people don't even make a significant effort to do this.

david - your second guess is closer to the truth. From a marginalist, individualist perspective, what's interesting is, first, whether or not the programs change behaviour. But if we're looking at, say, how the number of children people have responds to a change in Canada Child Tax Benefit - well, I suspect there probably is an effect at lower income levels, but in the absence of significant provincial variation in the CCTB, it's going to be pretty hard to tease it out of the data (I don't have time to merge multiple years of Canadian and US census data and try to do some kind of international difference in difference estimation, and wouldn't believe the "all else being equal" assumptions even if I did).

The incidence and take-up of the programs, which I can estimate, is only so intellectually exciting - I don't have the time it would take to get access the appropriate master files - and using those files is a real hassle - so I can't do a really fine grained incidence analysis. I'm mostly crunching Canada Revenue Agency's Income statistics, which is a very tedious task.

You might say none of this has anything to do with the social benefits of programs? That's true. But the person who tasked me with this assignment was using the term "social benefit" incorrectly, i.e. to refer to specific provisions of the income tax system, rather than to look at the social benefits of various policies.

Timlagor - A carbon tax is introduced, increasing the price of gasoline. What's your best guess as to what will happen to the amount people drive?

It's never wise to assume that people don't respond to incentives.

We are not just "social animals," with the trivial implication that we prefer company to solitude, like lonely puppies. No, as a species, we function socially. Our strength as a species lies in intricate cooperation (lateral meshing) and the management of resources over time (temporal meshing.) Lots of that meshing appears expendable or trivial, but we dismiss it at our peril.

To treat society and the economy as a given lake in which free agents pursue their individual goals, is to miss the essence of the problem altogether. Makes me nuts, listening to macro arguments that assume all the large and small chambers of society will just be there when wanted. This is not the case. Economics might not need to study this, I guess that their decision. After all, cooks don't need to study the electrical system that powers their stoves, or the soil chemistry that supports crops. But SOMEBODY has to.


Re: the Far Side... I guess because I am "old and female" (? sheesh, I'm only 38) I've seen both the dog and the cat ones, and I think the cat one is funnier, but they are good side-by-side:


I think that there is a lot to be mined in the areas of industrial organization, public health, transportation, and urban planning. Also studying productive eco-systems such as the rise of Detroit or Silicon Valley is very fascinating. I'm not sure how effective simple models would be in these areas. I know that agent based simulations are used in modeling things like traffic flow, but I don't think economists approve of these types of models.

Here is one of my favorite all time essays by Clay Shirky: "The Collapse of Complex Business Models", which I think provides a lot of valuable information of the socio-economic type, but would be difficult to model in a useful way.


@Frances Woolley:

> I call this the "social blah blah blah" phenomenon.

Maybe your conference paper can address exactly this? (And please with just this title?)

As confused as you are about just what "social benefits" means in the context of an income tax system, policymakers are even more confused. They don't have the economic foundation to precisely define their goals, leading to a political (and then popular) impression that the income tax system is some magic spell that can accomplish anything that the government wants.

Review bits of the income tax code to see what "social benefit" the government was apparently seeking, and then address whether the benefit delivered (or even could deliver) such a benefit.

I don't think this is about selective deafness in economists. Most people will view taxes as a cost, not a benefit, and while a cost may be incurred as the price of a benefit, how can it be a benefit in itself? The social benefit of the personal income tax system is pretty much that it gives us something to complain about at the pub.

Oh, and for a Far Side on social analysis, I'd go with "The Four Basic Personality Types" (https://www.pinterest.com/pin/8796161742129681/.)

Noni: "Our strength as a species lies in intricate cooperation (lateral meshing) and the management of resources over time (temporal meshing.)"

But you aren't using the word social here, are you? You're using very specific terms - e.g. intricate cooperation (lateral meshing) - to describe what sound to me like very specific concepts. If you had just said "social interaction" that wouldn't have been nearly so clear and precise, which presumably is why you didn't say it. You're not just saying social blah blah blah, you're trying to model the social - what I want economists to do.

rsj - interesting.

Majromax: "Review bits of the income tax code to see what "social benefit" the government was apparently seeking, and then address whether the benefit delivered (or even could deliver) such a benefit."

I think it'll be a subject for drinks in the bar afterwards... But, yes, I'm going to try to do something on program effectiveness. But to do that properly for programs for children/aged/disabled in 25 pages.... I could do one program, but not all of the ones listed. Also the stated intent of policy and the actual intent are not always the same...

Bean ;-). Thanks for adding the link.

Noni: I would say what (I think) you are saying this way: most of economics takes for granted a given system of (say) property rights and laws governing exchange, without which markets (and supply and demand curves) would not exist. But those rules and institutions (like property rights and markets) are themselves created by the interaction of individuals, and are (in principle) amenable to economic analysis.

Some (not very many) economists do indeed try to do just that. When I was young and keen and full of self-confidence I was one of them. Now I'm old and tired I just take them for granted, and just do macro.

Timlagor: about a month ago, I found myself in a restaurant in QUébec City, where I had gone last spring during another trip. In the middle of the meal, I realized I had ordered the exact same things as the last time. Creature of habit or rationnally making the same choice when faced with the same options?
As for the Far Side, two I remember is the one where the female pooddle's father ask the dog "So you work in the security department of a metal recycling firm?" and the one with the female monkey asking her husband about the blond hair.
Hope the long lnk work...) He sent Goodall the original as an apology and she was delighted...



Shorter link: http://ideonexus.com/wp-content/uploads/2011/09/JaneGoodallFarSide.gif


"I've been asked to write a paper reviewing social benefits delivered through the personal income tax system."

Compare and contrast other taxation models? Tax net worth rather than income? Rely on punitive taxation (fines)? Rely on charity?

Isn't this a loaded question? I don't think the question pertains to what tax revenue is actually spent on. Are there benefits (social, economic, political, other) to a government levying taxes even if government turns around and burns it's revenues? Do those benefits change based upon how taxes are assessed?

"Is social just a code word for not economics?"

I think it is easier to look at government policy through the other side of the lens (political) rather than social. How does a certain government policy (could be spending, could be taxation) benefit a politician responsible for implementing it?

Frank - "Compare and contrast other taxation models?"

Not in the time I have to write this paper!

"Social benefits delivered through the personal income tax system" -> support at the polls on election day. That's all it means - how do I redistribute to get enough votes to form a government. The word "social", in this context, is a shibboleth for "on the redistributionalist team".

I do not understand the reflex to defend the massive modern state. If you want to address grinding poverty for the externality that it represents, fine, it's simple, just do it. But let us remember that government has a monopoly on violence - so easily used to enrich a constituency and the well-connected while tyrannizing the poor and weak. Markets have no such power. Deirdre McCloskey puts it best:

"It would have been hard to know the wisdom of Friedrich Hayek or Milton Friedman in August of 1914, before the experiments in large government were well begun. But anyone who after the 20th century still thinks that thoroughgoing socialism, nationalism, imperialism, mobilization, central planning, regulation, zoning, price controls, tax policy, labor unions, business cartels, government spending, intrusive policing, adventurism in foreign policy, faith in entangling religion and politics, or most of the other thoroughgoing 19th-century proposals for governmental action are still neat, harmless ideas for improving our lives is not paying attention."

I am a bit disappointed that my remark about operational definitions, because they are the opposite of blahblahblah. It is no secret to anyone here that I am critical of the paucity of operational definitions in economics.

Oh, my! It's too early in the morning! I left out half a clause. :( I meant to say that I was disappointed in the lack of response to my remark about operational definitions.

Min - I didn't respond because I didn't feel like picking a fight.

You wrote: "Hmmm. When I read, "The analysis of social practice . . ." I think, it sounds like there are some operational definitions there. :)"

Here's the full quote: "The analysis of social practice constitutes a theoretically coherent and methodologically effective focus for social science research"

That's a statement *about* social practice, not a definition *of* social practice. Now I haven't read Fairclough, quite possibly there's a clear and coherent operational definition of what social practice is somewhere in his book. But that sentence I pulled out isn't one. Please do go and take a look. The quote is from Critical Discourse Analysis, chapter 7, section 4, and you can find the text on google books.

@ Frances Woolley

I don't think you are picking a fight. Me, neither. :)

Frances Woolley: "Here's the full quote: "The analysis of social practice constitutes a theoretically coherent and methodologically effective focus for social science research"

"That's a statement *about* social practice, not a definition *of* social practice."

Yes, and I am quite open to skepticism about whether the last part of the quote is justified. But social practices are things that can be operationalized, so that their analysis is well grounded. And, in fact, "social practice" is a far cry from things like "superego fragments", or whatever, which, IMO, are out in Lala Land. And now that you mention discourse analysis, it lends itself to well defined concepts such a minimum descriptive length and the cosine distance between texts. (Not that these are strictly objective measures. :))

I would be much more likely to have a blahblahblah reaction to statements such as these.

"The analysis of social norms constitutes a theoretically coherent and methodologically effective focus for social science research"

"The analysis of business confidence constitutes a theoretically coherent and methodologically effective focus for business cycle research"

Okay, I'll bite. Let me just give three examples of where "social" gets followed by a very non-economicsy word, but where the concept is extremely important, and the tendency to ignore it continues to significantly hamper economic analysis:

1. Social status. Status concerns are such an incredibly powerful motivating force -- particularly among academics! -- and yet economists have had so little to say on the subject. As academics, we implicitly understand this -- the status of our profession, and our institutions, is such that we are able to routinely extract millions of dollars from rich people, who feel that they have all the money they need, and yet not enough status. There are economists who have taken a serious crack at it (Veblen, Hirsch, Frank), yet anything resembling a model remains elusive. Partly that's because of the dynamics -- you want to write the status concerns into the utility functions, but they are constantly being changed by the interactions, so the standard game-theoretic equilibrium concepts don't work. But we ignore status at our peril. For instance, everyone knows that people derive a great deal of status from the amount of money they earn. And yet the status you get is determined by your pre-tax income, not your take-home pay. And yet how many economic models assume that people are solely interested in maximizing their post-tax income?

2. Social stigma. I once suggested that there is no such thing as a "rational choice criminologist." This is not strictly true, since there are criminologists who call themselves rational choice theorists, but if you take a look at what they are doing, they aren't really. Economics has, in general, done a great disservice to our understanding of crime and social deviance by putting disproportionate emphasis on the extrinsic incentives that people face (most obviously, legal punishment). With some this is ideological, but often it's just because these incentives are easier to model. If you look at what really discourages crime in any society, it's actually really diffuse, hard-to-model stuff, like the social stigma (or lack thereof) associated with being labelled a criminal, or the degree of attachment/alienation that a person feels towards the community. There's lots of hard data to back this up, but again, it's difficult to model action-theoretically.

3. Social norms. You mention this, but I have to point out that it took economists several decades and several million dollars spent on behavioural economics experiments to admit (in large numbers) that people often follow rules "just 'cause" -- and not for the extrinsic rewards associated with conformity/deviance. Again, there's no reason you can't model this (indeed, I wrote a book trying to show how you can do that). But it's important to acknowledge that the dominant response among economists, over the course of the 20th century, was not to say "how can I model this?" but rather to deny the phenomenon.

When people talk about "social benefits" they are typically referring to what we now call "network externalities." It is, however, important to recognize that this whole analysis of network effects is pretty recent, so that 30 years ago, when people talked about "social benefits" they were describing something real, that economists at the time didn't know how to model. So while you're right that hard core economists don't ignore the "social blah blah blah," but rather set out to model it (as they did in this case), one should also acknowledge that the profession has a really lamentable history of ignoring the blah blah blah. As far as modelling goes, economists have basically picked the low-hanging fruit, and most are (in my experience) content to pretend that the rest of the tree does not exist.

P.S. glad you liked my SEP article.
P.P.S. if you want to see someone taking a serious, no-bullshit crack at modelling "social practices," I would recommend Boyd & Richerson: http://www.press.uchicago.edu/ucp/books/book/chicago/C/bo5970597.html

Joseph - thank you for biting.

I think you and I are basically in agreement - things like social status do shape people's decisions, and economists don't pay as much attention to them as they should. The perverse effect of salary disclosure laws on executive salaries is a classic example. They were originally intended to keep salaries in check. But every CEO immediately checked out all the other CEOs' salaries, and immediately figured he was entitled to a raise - not because he wanted more money, but because he wanted more social status. A model of the firm that saw CEO salaries as related to CEO performance and determined by shareholders couldn't have predicted this effect.

Actually, as a university administrator, I find the Ontario salary disclosure list extremely handy, because I can figure out the relative status of all of the people in Central by checking out how they rank on the Sunshine List.

Having said that, I've just spent an enjoyable few minutes reading some of your latest blog posts. You might be surprised at how little you use the word social yourself. There's a tongue-in-check reference to "social classes" here. You quote Naomi Klein on social justice in your take-down of her book here. And this discussion of social costs and benefits is classic economic use of the words.

Perhaps this is just a small and unrepresentative sample of your writing, but I don't think so. In the three examples you give above - status, stigma and norms - the word "social" is unhelpful. For example, I care deeply about status - but is it meaningful, as Klein does in another context, to distinguish between the social and the economic? I don't think so. What matters, from a status point of view, is the reference group with which I'm comparing myself, and the key markers of status. It's possible to talk about status clearly and precisely without using the word social at all.

So discussions of human interactions, yes, a rich account of what gives our lives meaning and how we really make choices, yes, but social blah blah blah - not so much.

@Joseph Heath:

I think all of your concerns there translate into network effects and heuristics.

For both social status and social stigma, the concern is not a free-standing <blank>, it's <blank> relative to peers. If I earn more than my friends, then on a broad "lump of skill" level I must be more generally qualified. If productivity and income were directly correlated with IQ, then we really would want to pick the richest person to lead the country.

On the stigma front, whether I am a criminal or not doesn't matter in isolation, it's only in how people treat me. And here, what acts we consider "sufficiently criminal" to warrant concern depend on our social networks. Nobody cares about speeding and everyone cares about serial killing, but our stance on (say) drug crimes depends on our social network.

Social norms blend between the former and quantification of what we mean by "preference." Why do Canadians prefer hockey and Americans prefer baseball? Beats me, but once that preference is set then it is rational to go with the flow. The value from being a Leafs fan depends in part on how many fellow fans you have to regularly commiserate with.

I think that these ideas are precisely where "social" can be made "economic," through quantitative models of social networks. My intuition suggests a many-agent model, where social networks are imposed a priori but some variables (such as Pepsi versus Coke) are changeable, and utility functions relate to how many peers share a preference. I don't think that such models will nicely reduce to representative-agent analytical cases.

I'm a mathie and not an economist or social psychologist. Has modeling of this nature been done?

Frances and Nick, yes. I do try to model the social system, though I Am Not An Economist. Or perhaps because IANAE. The essence of understanding is to step back and forth between model and that which you're modelling, especially when your model keeps falling over. This is elementary, and I am certain there is no one in this conversation who doesn't know it. No egg-sucking lessons here.

God knows economics is complex and challenging, and mistakes will necessarily be made. But when modelling a complex system, one can set rather coarse limitations that allow you to move safely inside the parameters of necessity. I think of it like camping in the mountains. No matter how wide the ledge where you settle to sleep, you darn well make sure you don't roll out your sleeping bag at the edge. A just-in-time economy, which glorifies living on the edge, invites disaster.

Similarly, some elements of a society must be above, or below, certain sizes. Proportions matter a great deal*. And some "trace elements" must be preserved, even if they appear to be outdated or even pernicious.

Sadly, it seems that only SF authors pursue these topics in any depth. This study should be the core of policy, not a negligible frill.


* for instance, it appears that a 65-70% top marginal tax rate is the minimum needed to prevent an oligarch takeover.

Hi Frances,

Question 1) How did the requester of this paper on the personal income tax system define "social benefits"?

Question 2) How rigorous was this definition?

Comment A) If the definition is not rigorous, then the requester needs to redefine "social benefits".

Question 3) What is the level of agreement on this definition for "social benefits" in the requester's field?

Comment B) If there is a low level of agreement, then this field needs to agree on the definitions of its major concepts before doing any further work.

Question 4) In your observations, how pervasive is the use of non-rigorous definitions in the social sciences? In non-economics social sciences?

Comment C) As a statistician and a natural scientist, I commonly observe a lack of rigour and of agreement on key words in the social sciences. (Economists seem to be better about this, though I welcome feedback on this point.)

Comment D) Despite the interest and importance that many social science issues generate, those 2 limitations in Comment C) often make it hard for me to understand and converse productively with social scientists. Often, the discussions that I listen to or participate in become passionate monologues spoken by multiple people who happen to be in the same room, and no substantive progress was made from the discussion.

Comment E) I notice the problem in Comment C) to be even worse among academics who are in the arts or humanities but are not social scientists. Again, there may very well be some interesting or important issues to be discussed, and the people in those discussions may be very smart, but the lack of rigour and agreement in basic definitions prevent the discussions from becoming productive.

Comment F) Sometimes, if a definition is suffers from the limitations in Comment C), then the term is used in a politically convenient or favourable way for the user of the term. "Equality" and "Justice" are 2 examples; perhaps "Social" is another. (Read the NY Times article "Tyler Cowen on Inequality and What Really Ails America" to read how Tyler defines income inequality in 3 different ways.) Thankfully, an ionic bond or a standard deviation is not a politically charged concept, so I don't have to deal with this problem in my work.

Comment G) I hope that improvements can be made on the limitations in Comment C) and advance the dialogue and productivity of discussions in the social sciences. In particular, I wish that definitions are not exploited for political gains in intellectual discussions about important social issues.

Comment H) The word "social" connotes interactions with other people, yet there is much ambiguity about how large this circle of people should be. This circle may be the family, the neighbourhood, the city, the province, the country, the world......and the implications are quite different between these choices. I wish that social scientists would be more rigorous about the size and scope of this circle of people when they use the word "social".

Eric - here's the paper I'm supposed to write:

"1:45-2:45 pm: Delivery of Social Benefits Through The Personal Income Tax
-A review of social benefits currently delivered through the personal income tax system, including those for children (including for arts and physical fitness), persons with disabilities, the aged and others; and suggestion for any desirable reform."

Other than the social benefits stuff, it's pretty clear. And, truth be told, even though it's awkwardly worded, I know what they mean. As the deadline looms nearer - and standing up in front of all my tax econ buddies and making a fool of myself becomes a real and salient possibility - it is all coming sharply into focus!

Hi Frances,

It's good that *you* have a confident understanding of what they mean, but I think that it would still be a worthwhile exercise to ask them my Questions #1-3 above - precisely for the reasons outlined in Comments C-H above. I would encourage you to ask those questions without offering any options for them to choose from - this would obligate them to think carefully and rigorously about their own definitions, instead of benefiting from the labours of your thought process.

I recognize that interacting with one group of social science researchers is futile in the broad objectives of making definitions more rigorous and widely agreed upon in their fields. Nonetheless, this encouragement has to start somewhere, and this sounds like a good opportunity to do so, one research paper at a time.

I cannot see anyone being happy with this but it was pointed out to me some time back that everyone, including the poor, should not be exempt from taxation because it basically gives ALL taxpayers ownership in government. Social benefit could then be used to insist on universal tax paying. Not truly an economics statement but...

Just saw the invite to the conference - it's a pretty impressive lineup and a good collection of topics. I assume it's in Calgary (not obvious from the invite).

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad