[selmnoo]

> Just a few hours later, Vox published an article titled These tech interns are probably making more than you are where they shared a tweet showcasing candid salaries at various tech companies, some larger, some smaller:

That is a kind of a bullshit argument though isn't it, that "hey look, the interns (supposedly half-able, incompetent newbies) are getting paid so high!! We have such wonderful salaries in tech!"

It's a lie, a total misrepresentation of the facts. The interns are usually super programmers themselves. I mean, for instance, when Alex Gaynor was "interning" at Quora he ported Quora to run on PyPy. He's a really big contributor to projects like Django and PyPy. When he's "interning" at your company he's probably more able than a lot of other folk you could find, and he should be getting very big bucks. So this argument is really just complete hogwash. A good amount of times I see the interns training employees rather than the other way around (as it should be).

Secondly, they're interning in the valley, where "boarding" costs are about equal to costs of getting a mansion in a midwestern city in a reasonably safe and fun neighborhood. When you're "making 90k" in the valley you have the quality of life worth ~60k in a midwestern city. If anything, they are underpaid.

edit: minor revision on figures for clarity of argument; though I do agree with moc down below when he says "If you're comparing suburbs (Silicon Valley) to suburbs, the discrepancy is much greater and I'd say that $90k/CA ~= $40k/Midwest"

[dashboardfront]

> When you're "making 90k" in the valley you have the quality of life worth 30k in a midwestern city.

Not true at all, yet this is constantly repeated like it's fact.

Look up indifference curves and consumption bundles. You're ignoring microeconomics. For anyone that doesn't assign excessively high utility towards housing size, 90k in the valley is a much better quality of life than 30k in a midwestern city.

This is where cost of living calculators fail as well. You can't compare the same basket of goods everywhere you go. People will adjust their basket of goods depending on relative prices; it doesn't mean that their utility has dropped. It's income effect vs substitution effect.

Edit: The number above has been changed to 60k~. I'd say this is probably a more reasonable argument. As for the suburban argument, I don't really know the numbers and I can't argue there. As a person in my 20s who grew up in San Jose, the suburbs are like a bad dream that continuously haunts me.

[logfromblammo]

If you want your comparison to be equally valid wherever you make it, you have to use the same basket of goods everywhere you go. Otherwise, I can't tell whether you are comparing apples to oranges or pears to grapefruit.

When I look at the numbers for that basket of goods that I currently consume, and price it out for San Francisco, I find that it is completely unattainable for anything less than 3 times what I currently pay, and could be difficult to reliably source for less than a multiple of 4. That's the fact.

It does not matter that I would in reality have to substitute down to inferior goods. It would still be objectively worse than what I have now. The value of my willingness to substitute counterbalances the savings I make by substituting. If I move back to the cheaper area, I won't keep the SV basket; I will substitute right back to the best goods that I can afford.

[gjm11]

I agree that there's a big transparency benefit to using the same basket everywhere, but I don't think that actually gives the most realistic answer.

Imagine that every day, for every meal, you eat spaghetti carbonara. You'd be almost as happy eating linguine with pesto, but you happen to have a slight preference for spaghetti carbonara. Now you move somewhere else where, for whatever reason, spaghetti is 100x the price and every other ingredient is only 2x the price. Are you suddenly 100x poorer? No, much nearer 2x, because you'll just switch from spaghetti to linguine. You'll be slightly worse off than 2x because, darn it, you preferred spaghetti, but only slightly.

Moving to San Francisco from (say) Minnesota is a bit like that. Housing is a bajillion times more expensive, so you'll have to make do with a lot less of it, but unless housing is the only thing you care about that doesn't mean you're a bajillion times worse off. You do need some housing, and many other things are also more expensive, so there's no argument that you're worse off in SF for any given level of income. But not by the factor the price of housing would suggest.

(A couple of other remarks about this sort of comparison. 1. The richer you are, the less these things matter -- if you're putting a substantial fraction of your income into investments, those don't change in price at all just because you move to San Francisco. 2. Relatedly, if you are able to buy a house rather than renting, you're not as much worse off as the eyewatering price of the house would suggest -- because later on you can move out of San Francisco, sell that house, and get the money back again.)

[logfromblammo]

Realism takes a back seat to practicality here. A CoL calculator takes two cities and a dollar amount as inputs. A QoL calculator would need a lengthy list of preferences with reasonably accurate price estimates, and a complex network of likely substitutions. That seems like an academic project for a graduate student in economics, whereas CoL calculators are just a simple HTML form with a tiny bit of script and a database of scale factors for each city.

Thus, you can't really tell me how much I'd have to earn to be equally happy in SV as I am elsewhere, but you can easily tell me how much I'd need to maintain equal consumption there.

You can get some of your money back when reselling the house, possibly even within six months of listing it. I would not blithely make assumptions about any market when a bubble appears to be growing nearby.

[mbreese]

> because later on you can move out of San Francisco, sell that house, and get the money back again

You obviously didn't own a house in the US about 5-6 years ago...

[gjm11]

I obviously felt my comment was already too long and wouldn't on balance benefit from a cautious parenthesis to the effect that you might actually get a lot more money back than you put in, or a lot less, depending on exactly what the housing market does, and that it also depends a lot on how much of the house you own (the limiting case of a 100% mortgage is, at least to begin with, almost indistinguishable from renting), but that to first order, assuming a reasonable amount of equity and no catastrophic shocks to the housing market, what you get out is at least a fair fraction of what you put in ... but as two people have commented on that omission, perhaps I was mistaken.

(As it happens, I indeed didn't own a house in the US 5-6 years ago, but I'm well aware of what happened.)

[dashboardfront]

Yes, but using the same basket of goods is fundamentally flawed because people shift their consumption choices dependent on relative prices. People do not pick the same consumption bundles wherever they go. Your substitution to an inferior good (actually, just less consumption of a normal good, housing space) is compensated with a larger consumption of 'other goods', which have also rose in price but not nearly to the extent your salary went up (tripled, in his previous example). Again, income effect vs substitution effect.

[logfromblammo]

When you compare things, you have to limit the number of variables you change at one time.

[mbreese]

> Not true at all, yet this is constantly repeated like it's fact.

That's because it is true...

I moved from the Midwest to the peninsula 3 years ago (for my wife's work). I made roughly those numbers there and here and the standard of living was roughly the same, maybe even dropped a little. Living in Silicon Valley is easily 2-3X more expensive than living in a place like Indianapolis or Columbus, OH.

You honestly don't feel these pressures as much until you're older and have a family. At least, the differences between the regions are much more apparent with a family. For example, we need a certain amount of living space, and there are a set of fixed expenses. We lived in about the same size homes here and there, and just the cost for that alone more than tripled.

[dashboardfront]

Housing might very well be 2-3x (or more) expensive, but not all goods have gone up 2-3x more. This allows for an adjustment of your consumption bundle; your housing is not going to be as large, but you might have access to a greater amount of other goods. Of course, you can choose to easily make it 2-3x more expensive if you desire. Family makes this difficult because your personal indifference curve places higher utility towards space, but the parent was discussing intern salaries, who generally don't have to support families.

[mbreese]

You have the same access to goods almost anywhere in the US. The basic problem is that certain basic needs (energy, housing, child care) cost 2-3X more in the Bay Area than the Midwest. The only things that isn't the case for are food and transportation, which are slightly more expensive, but not as nearly as much. Transportation may be slightly cheaper out here, but not by a significant margin.

Your idea of a consumption bundle is appealing, but not very practical... you are assuming that there are things that can be adjusted, but for many, that just isn't the case. If you don't have a family, you might be able to get away with living in a smaller place. But if you do have one, for any given standard of living, there is a certain minimum size that you need. And the increase in costs in housing severely limit the flexibility that you might have in other areas.

I was mainly replying to your assertion that it isn't 2-3X more expensive to live in the Bay Area than the Midwest. I'm sorry to tell you that it is.

[dashboardfront]

The idea of a consumption bundle is a microeconomic theory that has a lot of evidence behind it and is generally accepted in the community. It's not really my idea.

There is no certain minimum size that you need; you simply place a lot of utility of living space as opposed to other goods up to a certain point, at which diminishing returns kick in. The amount of utility you place clearly increases as you have more family members to take care of. In the case of an intern, this is generally not an issue.

There are also other purchases besides food and transportation that do not increase; the cost of travel and vacationing, for instance. Technology generally remains the same cost throughout the US. Clothing, I'd imagine, doesn't increase too much.

> I was mainly replying to your assertion that it isn't 2-3X more expensive to live in the Bay Area than the Midwest. I'm sorry to tell you that it is.

It can be. It depends on the utility you place on sq ft as opposed to all other goods. However, it is not an axiom that one needs 3x the money to live in the Bay in order to remain at the same indifference curve level as they previously did in the Midwest.

[krschultz]

Exactly. When I lived in the suburbs I had a 3 bedroom condo, and my wife and I owned two cars. I moved to the city and we now have a 1 bedroom apartment and no cars. Total cost of living is roughly the same. My happiness in my living arrangements is about the same. A cost of living calculator would make it look like we are getting absolutely hosed.

Our take home $ into bank accounts tells an entirely different story. The raise more than made up for any increased cost of living.

It kind of makes sense too, if it didn't make sense, people wouldn't be doing it, there wouldn't be so much demand, and prices would fall.

[jarek]

You can also live in the city with no cars in midwest or many other places with lower CoL than Bay Area

[UK-AL]

Surely most of you money is going on rent, so the money left will be the same as other cities.

[dashboardfront]

You can always adjust your consumption bundle. If I'm living in a high rent city, I'm more likely to find a friend to share an apartment with or settle for something not that great. This is a drop in utility (though I'd argue living with a friend is a positive for myself), but I also have a lot more money to use on other things, which increases my utility. Spending most of your money on rent when you're pulling something like 5,000 after tax a month is a choice, not a given, and is reflective on how much utility you perceive a better apartment to give you. In reality, we're probably terrible at accurately estimating utility of a good, but I personally have lived in a variety of places, and I was most happiest living in my tiny dorm room, so I personally don't care much about having a big place.

[UK-AL]

Your average friend is likely to want a close to market price for that room. Relying on friends is not a way to plan an economy. A lot of people won't have friends willing to let them stay.

[dashboardfront]

That's not what I was saying at all, but alright. If you're interested in the topic, look up utility curves, income effect, and substitution effect in a microeconomics textbook. It'll probably have a much better explanation than I can communicate through text (unless I start making graphs, which would be a bit time consuming).

Here's a copy paste of my explanation on reddit though (where this topic also constantly comes up):

Living in the Bay Area results in a higher wage, which increases consumption (income effect), but at the same time the price of housing is much higher as well (to a greater degree than the increase in income). You substitute sq ft with other goods. If you place relatively heavy utility on housing by sq ft in comparison to all other goods, then I guess it's a bad idea to come out here, because you're going to be on a lower indifference curve. Otherwise (and I think most people fall under 'otherwise', which is why they come out here ) its not that bad of an idea, and you'll probably end up on a higher indifference curve.

[UK-AL]

I think you need a certain amount of sq ft regardless. Most people aren't going to trade living in a tiny room for more stuff.

Your living area is probably No.1 contributor to quality of life.

[michaelochurch]

If you're comparing cities to cities (e.g. Chicago vs. SF, Minneapolis vs. Seattle) then I'd say that it's closer to 65k than 30k to get an equivalent lifestyle to $90k on the West Coast. There are (contrary to coastal stereotype) great places to live in the Midwest, and they are cheaper than SF, but they aren't super-cheap either. The 2500 SF houses for $300k aren't in downtown Chicago. They're in suburbs where most 25-year-olds would be pretty bored because walking or biking to work is very rare and you can't go out without a designated driver (i.e. cab service and public transportation aren't options).

If you're comparing suburbs (Silicon Valley) to suburbs, the discrepancy is much greater and I'd say that $90k/CA ~= $40k/Midwest. That's because Silicon Valley has so much less in the way of urban amenity, but still charges an urban premium that you wouldn't see 20 miles out of Minneapolis. When you adjust for commute, suburban California ("the Valley") looks even worse.

[drivingmenuts]

Purely on the informative side: why are these people called "interns" when it looks more like they're straight-up "contractors" or "employees"?

Is this just a convenient fiction?

[teacup50]

Most of them are worthless; of any given intern class at a first tier technology company (e.g., Apple), only ~1-5% will actually be offered a full time job.

Those 1% are worth every penny, but you simply don't know who they are until you give the entire intern class the opportunity to demonstrate their abilities.

[selmnoo]

> Most of them are worthless

They are almost never worthless. Small startups who are not experienced with interviewing may get a few bad apples, but companies like Quora (and the others on the list in submission article) don't make those mistakes. Sure, they may not all be Alex "Im-a-core-dev-for-Pypy" Gaynor's, but generally they'll be worth every penny of that 90k and another 90k more.

[teacup50]

I expect that webdev shops have lower skill bars than OS development shops.

[pauleastlund]

In my experience at Google, far, far more than 1-5% were offered full time jobs. Maybe my group just had above-average interns? But I don't think so. The intern hiring bar was fairly high.

[potatoyogurt]

Yeah, this is silly. The point of having interns is like 90% finding and vetting new employees. Certainly, not everyone will be offered a job, but if less than 5% of a company's interns are receiving offers, there is something seriously wrong with the company's intern selection process.

[jpatokal]

Do you have some sort of source for that figure? Every company I've worked at, including both Google and obscure outfits you've never heard of, has made an offer to far more than 1-5% of its interns. This random Quora answer says 70% at Google get offers, which seems high to me, but less off than 5%.

http://www.quora.com/What-percentage-of-Google-Interns-get-f...

[michaelochurch]

Most of them are worthless; of any given intern class at a first tier technology company (e.g., Apple), only ~1-5% will actually be offered a full time job.

I really doubt this number. Rehire rates are an important consideration for college students, and they talk. An offer rate below 50% would be damaging.

That said, a lot of people find better offers elsewhere, but no tech company is going to no-offer 90% of its interns and keep its reputation intact.

[yurisagalov]

actually, I am not claiming at all that interns are half-able, incompetent newbies. Some of our best work has been done by people interning with us, and I think our past interns will vouch for that. That being said, it is true that intern salaries are generally lower than full time salaries (although the gap is shrinking, IMO), and intern salaries are good indicators of full time salaries in tech.

[anthony_d]

The thing is that a highly paid intern in no way refutes the argument that immigration is about cheap labor. The argument is that 1) because domestic labor is expensive companies want to increase H1B's as a way to decrease cost 2) if you didn't mind paying more you could find domestic labor. The highly paid intern is an point in favor of #1.

Nothing in the article disputes the basic economics of #2. Sure there is competition for talent, there is competition in every facet of business. The trick is to find the ratio between how much value a business extracts from a resource and how much value the business provides to the resource; that ratio needs to stay above 1.