How many of those people who die and are injured because people think some jackass should be free to hop in their car drunk and kill someone? If you could be sure drunk drivers would only kill themselves then sure, evolution at work. But when the victims of such 'freedoms' are innocent bystanders....
Depends how they calculate 4 billion dollars. If their entire itanium business reseted on Oracle continuing support and the itanium business was 4 billion dollars then it's not crazy.
It seems crazy. But this is a lot of expensive stuff we're talking about. The hardware is cheap, it's the software, and importantly the data in the software that matters, and that's big money. Migrating it all to another setup could be expensive, replacing all the hardware is expensive etc.
Except that the curve (not my choice for a fit) is chosen specifically because it hits most of them most closely, that's why you'd use linear regression at all.
But you do need bigger data sets. And a single outlier on a large dataset (again, not my dataset), that is from a single test is impossible to ascribe any particular significance to. You'd need to know the error rate at each data point, repeat the test several times etc.
And the problem with that plan is that if it gets published it gets paywalled.
The second link *should* have error bars on their data points, with 55 and 77 people there are enough to actually put in the variability, even if it's crude. But well... you can only expect so much of medical doctors (who are not scientists generally).
Sorry, I guess I didn't draw the connections well.
For the services you get you need to pay whatever percent in tax. On average you pay some current percent in tax (regardless of occupation). Since IT is not a particularly exceptional position the overall average data more or less applies. Expect to be taxed at roughly the rate given.
The deficits only come into play because you can't just compare tax rates. The current US effective tax rate of 27% is way off from it's outlays, so it's obviously anomalous. Either the US will have to increase effective tax rates (which might mean just putting more people to work to increase the tax base) to match its spending eventually, or it will have to massively cut spending. Germany is in the same boat on a smaller scale.
Perhaps more simply: Tax rates are a bad measure because in the US you aren't paying for what you're getting, nor is germany, but they're much closer to actually paying for what they get.
Now I grant you that total tax as a percent of GDP is an awkward measure. But it's a good overall average estimate, because otherwise you'd have to do a very complex analysis of various income levels in various locations throughout germany and the US and sum up the total tax burdens in each place. While I could certainly try and do that, I don't have the data in front of me so it's a non trivial amount of research to even do the analysis. The chart I linked is a pretty good 'overall average' if you will. Trying to pick say 1000 locations in each place, at 3 or 4 different reasonable IT incomes, to sort through all the local tax laws, state tax laws, federal tax laws, costs of living etc. is not something I have time to do for a/. post (and if I had time to do it elsewhere I'd just link to it).
Not when I'm actually being an intellectual. What fun are forums if you can't troll people who say obviously clueless things?
He was making a clearly intellectually dishonest statement at someone who suggested the drinking limit should be 0, I saw no reason to change the tone to convey the point that his response was clueless.
Someone else (replying to the same OP) presented a much more serious statement in a much more serious way, that warranted a researched and thought out response with linked to actual academic studies. To that end, that discussion touches on more serious issues like the perception of alcohol impairment, risk tolerance and so on, and warrants a more.
Also, your signature exists to troll people in the Steve Jobs reality distortion bubble (whether on topic or not), and despite the year lacks the broader context of what was said. So how is that fundamentally much different from what I did? These are forums, not scientific conference discussions, and even in those you get to troll the crazy people.
Worst case that's your pension. It's still a pension. Shitty as it may be. Although if you follow the paperwork there, there are a few additional avenues of getting stuff, food stamps, medicare supplements, medicaid, the social security you earned (on top of SSI).
But that's why governments hammer away on you when you're sober about what the limit is. In the hopes that you'll remember enough that when you're drunk you won't drive, or that someone sober will step in and stop you.
I did. And you were decidedly unsubtle about your analysis, so I felt no reason to provide a sophisticated retort. I saved that for the idiot below you.
LOTS of countries have taken the 'no measurable amount' as the tolerance limit by the way. As I say in reply to someone below you there's no good evidence for that, 0.04 is not meaningfully much different.
Sure, though 0.04 is something like one-two drinks within an hour for an average sized person kind of thing. I would think that's why the threshold is set at 0.05 most places, because a small person with a big glass can go over the limit a bit, and not be hugely worse off.
The Japanese take a different approach, which is basically don't be impaired. Drunk, tired, high, they don't care. They can basically pull you over for being in any way impaired. That seems a bit too unscientific for my taste.
Essentially a thin woman (~45 kilos) will add 0.05 per drink, so they're just at the point of having measurable effects for about an hour after one drink. http://en.wikipedia.org/wiki/Blood_alcohol_content
Backed up by what? This is where you need fairly sophisticated economic analysis to judge just how expensive things become.
The question you're asking is what is the cost of increased accidents versus the benefits? If a 0.05 BAC increases accidents by X a year with Y cost, and you have Z in benefit from extra economic activity due to more drink sales, is Z> Y, is there a moral cost to the added accidents etc.
I didn't actually apply a standard. hard to fathom here I know, but I didn't. There's almost no measurable impairment at 0.04. After that the 'standard' I applied was that you have to do a risk calculation and decide risk tolerance. Obviously you and every government in the world disagree.
The first chart in the first link is illustrative. A drop from 1 to 0.96 more or less corresponds to 0.05 BAC, and down to 0.93 ish is 0.08. But that doesn't tell you what happens to driving accident rates. Quite the contrary, it's a purely synthetic test.
a sample for people wondering about actual data. http://www.ahw.org.nz/resources/pdf/AHWbriefingpaper_2003-1_BAC.pdf from New Zealand (population approximately 4 million).
Based on overseas experience, by lowering the legal blood alcohol concentration from 80mg to 50mg we could expect between 16-72 lives saved, and 640-1280 injuries avoided each year.
As buried within that document they are trying to get drunk driving deaths below the arbitrary threshold of 300.
So pretending you're from new zealand. How much is 16-72 lives and 640-1280 injuries worth, compared to the economic benefit of increased drink sales and cab rides? How do you evaluate the social benefits and harms?
They shouldn't, at least not from their employers. Why would you want to be trapped in one job for your whole life?
Depends where you live. You can transfer pensions between employers here. Some places have government 'earned pensions' where you pay into the government pension system and the more money you paid in the more you collect when you retire.
Those are still pensions.
The other thing that happens here is that you can collect a pension when you retire from all of your past employers. So if you worked for 10 employers you get 10 pension cheques.
Agreed. There's a risk tolerance question. I posted in reply to someone below (but in reply to the same OP) something a bit more sophisticated. There's no meaningful dropoff in performance up to about 0.04 BAC, after that everyone starts to perform worse more or less equally. Beyond that it's really a matter for politicians to decide how much risk and what cost benefit is appropriate.
The most likely occurrence is certainly the drunk careening into a parked or stopped car. Although I have been in a bank when a drunk crashed through the door in the middle of an afternoon.
I actually did a more sophisticated analysis, I was trying to be illustrative not exact. It's posted above if you want to read it.
he wouldn't be getting a loan, he'd be stuck paying the fine, now whether or not sweden has interest on its fines I don't know.
"Legal interest" is the maximum amount one can legally charge in interest. I'm not sure that applies here, since again, it's not a loan.
As it turns out, my bad assumption was that the average income for a swede over a 30 year period of 100k is too low based on the last 30 years. It should be 126K or thereabouts. But then my assumption that he could pay 50% was wrong, because I think the way the EU does poverty calcs is that he's be limited to about 40% of median income as debt repayment.
I agree that I haven't factored in taxes. That's actually tricky, because poverty numbers are calculated before tax, and paying off a legal fine might be tax deductible (if it's garnished from his wages).
The interest thing in general is tricky. 12% interest and sure, he's never going to pay it back. But even if he can get an exactly average job they can't garnish his wages below the poverty line (which is I think in the EU 60% of median wage), which is about 40k USD a year. That's not spectacular, but you can have a decent enough life on that and still get it paid off if it's low enough interest.
uh.. if you're 34 years old you have two choices. Leech off relatives for the next 30 years. Or work somewhere for the next 31 years until retirement, and what lifestyles would you have in either case.
Such as figure 1B, which shows, on average, how mean relative performance always decreases with any alcohol (albeit trivially up to about 0.04)
or http://addictions.uchicago.edu/carl/DandAlcDependence%20Brumback.pdf
That shows only with low doses of alcohol (under about 0.042 BAC) can you not really notice a drop in performance, and after that everyone, heavy drinkers or not perform worse in cognitive tasks and that the heavy and non heavy drinking groups mirror each other in performance.
I'd give more links but if you aren't on a university campus or somewhere else that they're free it's sort of futile, and I don't really want to keep mashing links until I find ones that work without academic library access (and where I am has a med school so we're subscribed to medical journals, places without a med school might not).
But hey, why not make up some facts that 'some studies' support your argument so you can create a perception of authority without providing those studies and when the most easily findable studies (when searching for terms like cognitive impairment and blood alcohol level, and other obvious search terms produce results that disagree with you?).
Now your first line, you hit the nail on the head, before you joined the nutter bin. It's clear that BAC less than about 0.04 has so little impairment (even though it is there) you need very large sample sizes for that effect to not get lost in noise. That applies to heavy and mild drinkers equally. So if where you live has a BAC requirement of less than 0.04 they're probably playing theatre or moral/religious grounds than evidence based policy. Anyone who's at about 0.05 or worse is making a judgment call on just how much measurable impairment is tolerable, which is all science can do. Politicians have to decide risk tolerance and cost benefit analysis, science merely quantifies the effect that creates risk, and sometimes the risk itself.
Say that when a drunk crashes through your front door, or hits your parked car, or runs into your kid on the side walk, or hits an electrical box in front of your house. Should all those people have been free to decide whatever too? How about when you, with your perfect knowledge of how to drive is stopped at a red light and a drunk careens into the back of your car?
Sort of by definition if your judgment is impaired you're not capable of making a judgment about when its safe to drive. Governments then take the view that at some point the average person becomes sufficiently impaired that they cannot be safely on the road, and then hammer this point into you when you're sober in the hopes that you'll remember when you're drunk (or that someone sober will keep you from killing yourself or someone else).
It's the government being nanny state douchebags telling you what to do, and it's not doing enough to keep stupid people off the road when they crash into you and yours.
3. I arbitrarily limited to 30 years to convey a point.
4. Everywhere in the west have minimal old age pensions for people with low income.
Lets do the maths again. The guy in question is Fredrik Neij, a 34 year old swede.
Average monthly salary in sweden is 3,911 (wiki page on sweden) euros, or 5,279 dollars. X12 = 63000 dollars.
I'm not entirely sure how poverty is calculated in Europe or sweden specifically. The UK uses 60% of median income as it's poverty rate, so lets stick with that measure, because you'd think that would be consistent with the EU.
By that account one would need 38k (more or less) dollars per year to not be in 'relative' poverty. Before tax. Leaving 25k per year (admittedly, before tax) to be used to pay off debt. So it would take 60 years, but he's still be living on just under 40k equivalent his entirely life. That seems not great but manageable. Except we haven't taken into account wage growth over time.
Which takes us to to a question of 'what is his average salary going to be over 30 years'. That is a much harder question.
http://www.scb.se/Pages/TableAndChart____28313.aspx gives a longer more complex breakdown (that doesn't quite agree with wikipedia above but lets roll with it for the moment). In 1982 the 'average' salary in sweden was about 9000 Euro, compared to about 40k Euro now. Roughly a factor of 4 increase over 30 years. If you (wrongly but for sake of simple argument) assume that's linear. Then you can expect that in 15 years the median salary will be 126 dollars equivalent which would be the 'average' salary for the whole 30 year period.
So actually I undershot the projected income, but overshot on how much he'd keep. And I stand by my 30 year assessment then. It would take 15 years to be paying 50k (almost exactly, conveniently enough), but every year after that, assuming he only ever earned the exact average salary of an swede, he'd be paying more than 50k and have it paid of with 1 year before retirement at 65.
On top of that remember that's only his personal incomes. If he got married to someone also making the exact average income (and managed to stay together) they'd be doing pretty well for themselves.
So lets revisit what you said.
A normal person makes far less money than that. Also normal people do not get pensions. Those went away decades ago.
First part: True for today. Not true for the average swede over a 30 year period. second part: not true for sweden at all. Not true for most decent places to live.
So given the choice between leeching off relatives until you die or living on the relative poverty line of 60% of median income (which is apparently about 38K USD right now) until retirement which would you prefer?
Obviously if he has to pay interest on that debt he's in trouble unless he can get a million dollar book deal. But a million and a half dollars is possible to pay off if you're in your 30's.
Again, I'm assuming he makes exactly the average income, if he manages anymore than that then it becomes significantly easier.
1.5 million you could pay back if you had a decent enough job. 100k a year minus 50k for 30 years. And you'd still collect pension benefits as though you had 100k.
That's manageable for a normal person. 1.5 billion... not so much.
Also, living with family is risky, because any assets you get in inheritance will still be claimed, and any needs you have as you get older will not be covered if your siblings/relatives can't, and you won't have any pension except government security.
oh and interestingly, note on that that UCLA has a "Nonresident Supplemental Tuition* 15,102.00" Which is I'm guessing the out of country/state fee but only applies to masters students.
Comp sci, physics and engineering. Oh and a neuroscientist. All have to pay tuition. And I can count about 30 in the US that I know personally, and we see quite a number (10 or 15 a year) here looking at grad school because they can't afford tuition in the US in a comp sci department.
Admittedly some of this is taking money from one pocket to pay another. I get paid in part based on the assumption that I have 7K in tuition, so that's budgeted as part of how much I get paid, but I still have to pay it. Foreign students however don't get paid differently than I do, and they have to pay 3x my tuition.
E.g. UCLA
http://www.registrar.ucla.edu/fees/gradfee.htm
Tuition 11,220.00
For grad students. Now a domestic (US) student would be paid enough that presumably they can cover some or all of that and still live.
Not that much more. A PhD in physics, chem or maths is mostly in the professor bracket. Which typically cap out in the 160k range as a full prof with 30 years experience. But you start at 50-70 and claw your way up from there. Engineering is a bit more than that, but not a lot. You can make more than that in industry of course, but the vast majority of PhD's in science are still academic type jobs.
Slashdot Mirror
Hah. Ya, it's sad people think that way.
How many of those people who die and are injured because people think some jackass should be free to hop in their car drunk and kill someone? If you could be sure drunk drivers would only kill themselves then sure, evolution at work. But when the victims of such 'freedoms' are innocent bystanders....
http://www.cdc.gov/MotorVehicleSafety/Impaired_Driving/impaired-drv_factsheet.html
180 children... (admittedly that's about 2% of the drunk driving deaths, but still ugh).
And that is with drunk driving rates being significantly lower than in the past.
Ya. Most of the world it's 0.05 but the UK, canada and the US are still at 0.08 which is where most countries set it in the 70's or so.
Depends how they calculate 4 billion dollars. If their entire itanium business reseted on Oracle continuing support and the itanium business was 4 billion dollars then it's not crazy.
It seems crazy. But this is a lot of expensive stuff we're talking about. The hardware is cheap, it's the software, and importantly the data in the software that matters, and that's big money. Migrating it all to another setup could be expensive, replacing all the hardware is expensive etc.
Except that the curve (not my choice for a fit) is chosen specifically because it hits most of them most closely, that's why you'd use linear regression at all.
But you do need bigger data sets. And a single outlier on a large dataset (again, not my dataset), that is from a single test is impossible to ascribe any particular significance to. You'd need to know the error rate at each data point, repeat the test several times etc.
And the problem with that plan is that if it gets published it gets paywalled.
The second link *should* have error bars on their data points, with 55 and 77 people there are enough to actually put in the variability, even if it's crude. But well... you can only expect so much of medical doctors (who are not scientists generally).
Sorry, I guess I didn't draw the connections well.
For the services you get you need to pay whatever percent in tax. On average you pay some current percent in tax (regardless of occupation). Since IT is not a particularly exceptional position the overall average data more or less applies. Expect to be taxed at roughly the rate given.
The deficits only come into play because you can't just compare tax rates. The current US effective tax rate of 27% is way off from it's outlays, so it's obviously anomalous. Either the US will have to increase effective tax rates (which might mean just putting more people to work to increase the tax base) to match its spending eventually, or it will have to massively cut spending. Germany is in the same boat on a smaller scale.
Perhaps more simply: Tax rates are a bad measure because in the US you aren't paying for what you're getting, nor is germany, but they're much closer to actually paying for what they get.
Now I grant you that total tax as a percent of GDP is an awkward measure. But it's a good overall average estimate, because otherwise you'd have to do a very complex analysis of various income levels in various locations throughout germany and the US and sum up the total tax burdens in each place. While I could certainly try and do that, I don't have the data in front of me so it's a non trivial amount of research to even do the analysis. The chart I linked is a pretty good 'overall average' if you will. Trying to pick say 1000 locations in each place, at 3 or 4 different reasonable IT incomes, to sort through all the local tax laws, state tax laws, federal tax laws, costs of living etc. is not something I have time to do for a /. post (and if I had time to do it elsewhere I'd just link to it).
warrants a more thorough analysis.
If not proofreading.
Not when I'm actually being an intellectual. What fun are forums if you can't troll people who say obviously clueless things?
He was making a clearly intellectually dishonest statement at someone who suggested the drinking limit should be 0, I saw no reason to change the tone to convey the point that his response was clueless.
Someone else (replying to the same OP) presented a much more serious statement in a much more serious way, that warranted a researched and thought out response with linked to actual academic studies. To that end, that discussion touches on more serious issues like the perception of alcohol impairment, risk tolerance and so on, and warrants a more.
Also, your signature exists to troll people in the Steve Jobs reality distortion bubble (whether on topic or not), and despite the year lacks the broader context of what was said. So how is that fundamentally much different from what I did? These are forums, not scientific conference discussions, and even in those you get to troll the crazy people.
In the USA pensions went away a long time.
http://www.ssa.gov/pubs/11000.html
Worst case that's your pension. It's still a pension. Shitty as it may be. Although if you follow the paperwork there, there are a few additional avenues of getting stuff, food stamps, medicare supplements, medicaid, the social security you earned (on top of SSI).
Not that wide really no. About 1 drink.
But that's why governments hammer away on you when you're sober about what the limit is. In the hopes that you'll remember enough that when you're drunk you won't drive, or that someone sober will step in and stop you.
Naturally. this was in reply to various posts about where the blood alcohol limit is set.
The question of TFA and what standards of evidence buggy software need to meet are very different and far more complicated.
I did. And you were decidedly unsubtle about your analysis, so I felt no reason to provide a sophisticated retort. I saved that for the idiot below you.
LOTS of countries have taken the 'no measurable amount' as the tolerance limit by the way. As I say in reply to someone below you there's no good evidence for that, 0.04 is not meaningfully much different.
Sure, though 0.04 is something like one-two drinks within an hour for an average sized person kind of thing. I would think that's why the threshold is set at 0.05 most places, because a small person with a big glass can go over the limit a bit, and not be hugely worse off.
The Japanese take a different approach, which is basically don't be impaired. Drunk, tired, high, they don't care. They can basically pull you over for being in any way impaired. That seems a bit too unscientific for my taste.
Essentially a thin woman (~45 kilos) will add 0.05 per drink, so they're just at the point of having measurable effects for about an hour after one drink. http://en.wikipedia.org/wiki/Blood_alcohol_content
That would be your risk tolerance calculation.
Backed up by what? This is where you need fairly sophisticated economic analysis to judge just how expensive things become.
The question you're asking is what is the cost of increased accidents versus the benefits? If a 0.05 BAC increases accidents by X a year with Y cost, and you have Z in benefit from extra economic activity due to more drink sales, is Z> Y, is there a moral cost to the added accidents etc.
I didn't actually apply a standard. hard to fathom here I know, but I didn't. There's almost no measurable impairment at 0.04. After that the 'standard' I applied was that you have to do a risk calculation and decide risk tolerance. Obviously you and every government in the world disagree.
The first chart in the first link is illustrative. A drop from 1 to 0.96 more or less corresponds to 0.05 BAC, and down to 0.93 ish is 0.08. But that doesn't tell you what happens to driving accident rates. Quite the contrary, it's a purely synthetic test.
a sample for people wondering about actual data. http://www.ahw.org.nz/resources/pdf/AHWbriefingpaper_2003-1_BAC.pdf from New Zealand (population approximately 4 million).
Based on overseas experience, by lowering the legal blood alcohol concentration from 80mg to 50mg we could expect between 16-72 lives saved, and 640-1280 injuries avoided each year.
As buried within that document they are trying to get drunk driving deaths below the arbitrary threshold of 300.
So pretending you're from new zealand. How much is 16-72 lives and 640-1280 injuries worth, compared to the economic benefit of increased drink sales and cab rides? How do you evaluate the social benefits and harms?
They shouldn't, at least not from their employers. Why would you want to be trapped in one job for your whole life?
Depends where you live. You can transfer pensions between employers here. Some places have government 'earned pensions' where you pay into the government pension system and the more money you paid in the more you collect when you retire.
Those are still pensions.
The other thing that happens here is that you can collect a pension when you retire from all of your past employers. So if you worked for 10 employers you get 10 pension cheques.
Agreed. There's a risk tolerance question. I posted in reply to someone below (but in reply to the same OP) something a bit more sophisticated. There's no meaningful dropoff in performance up to about 0.04 BAC, after that everyone starts to perform worse more or less equally. Beyond that it's really a matter for politicians to decide how much risk and what cost benefit is appropriate.
Trying to be dramatic requires being dramatic...
The most likely occurrence is certainly the drunk careening into a parked or stopped car. Although I have been in a bank when a drunk crashed through the door in the middle of an afternoon.
I actually did a more sophisticated analysis, I was trying to be illustrative not exact. It's posted above if you want to read it.
he wouldn't be getting a loan, he'd be stuck paying the fine, now whether or not sweden has interest on its fines I don't know.
"Legal interest" is the maximum amount one can legally charge in interest. I'm not sure that applies here, since again, it's not a loan.
As it turns out, my bad assumption was that the average income for a swede over a 30 year period of 100k is too low based on the last 30 years. It should be 126K or thereabouts. But then my assumption that he could pay 50% was wrong, because I think the way the EU does poverty calcs is that he's be limited to about 40% of median income as debt repayment.
I agree that I haven't factored in taxes. That's actually tricky, because poverty numbers are calculated before tax, and paying off a legal fine might be tax deductible (if it's garnished from his wages).
The interest thing in general is tricky. 12% interest and sure, he's never going to pay it back. But even if he can get an exactly average job they can't garnish his wages below the poverty line (which is I think in the EU 60% of median wage), which is about 40k USD a year. That's not spectacular, but you can have a decent enough life on that and still get it paid off if it's low enough interest.
uh.. if you're 34 years old you have two choices. Leech off relatives for the next 30 years. Or work somewhere for the next 31 years until retirement, and what lifestyles would you have in either case.
http://www.fatiguescience.com/assets/pdf/Alcohol-Fatigue.pdf
Such as figure 1B, which shows, on average, how mean relative performance always decreases with any alcohol (albeit trivially up to about 0.04)
or http://addictions.uchicago.edu/carl/DandAlcDependence%20Brumback.pdf
That shows only with low doses of alcohol (under about 0.042 BAC) can you not really notice a drop in performance, and after that everyone, heavy drinkers or not perform worse in cognitive tasks and that the heavy and non heavy drinking groups mirror each other in performance.
I'd give more links but if you aren't on a university campus or somewhere else that they're free it's sort of futile, and I don't really want to keep mashing links until I find ones that work without academic library access (and where I am has a med school so we're subscribed to medical journals, places without a med school might not).
But hey, why not make up some facts that 'some studies' support your argument so you can create a perception of authority without providing those studies and when the most easily findable studies (when searching for terms like cognitive impairment and blood alcohol level, and other obvious search terms produce results that disagree with you?).
Now your first line, you hit the nail on the head, before you joined the nutter bin. It's clear that BAC less than about 0.04 has so little impairment (even though it is there) you need very large sample sizes for that effect to not get lost in noise. That applies to heavy and mild drinkers equally. So if where you live has a BAC requirement of less than 0.04 they're probably playing theatre or moral/religious grounds than evidence based policy. Anyone who's at about 0.05 or worse is making a judgment call on just how much measurable impairment is tolerable, which is all science can do. Politicians have to decide risk tolerance and cost benefit analysis, science merely quantifies the effect that creates risk, and sometimes the risk itself.
Say that when a drunk crashes through your front door, or hits your parked car, or runs into your kid on the side walk, or hits an electrical box in front of your house. Should all those people have been free to decide whatever too? How about when you, with your perfect knowledge of how to drive is stopped at a red light and a drunk careens into the back of your car?
Sort of by definition if your judgment is impaired you're not capable of making a judgment about when its safe to drive. Governments then take the view that at some point the average person becomes sufficiently impaired that they cannot be safely on the road, and then hammer this point into you when you're sober in the hopes that you'll remember when you're drunk (or that someone sober will keep you from killing yourself or someone else).
It's the government being nanny state douchebags telling you what to do, and it's not doing enough to keep stupid people off the road when they crash into you and yours.
1. I said decent enough job.
2. I didn't include inflation, investments etc.
3. I arbitrarily limited to 30 years to convey a point.
4. Everywhere in the west have minimal old age pensions for people with low income.
Lets do the maths again. The guy in question is Fredrik Neij, a 34 year old swede.
Average monthly salary in sweden is 3,911 (wiki page on sweden) euros, or 5,279 dollars. X12 = 63000 dollars.
I'm not entirely sure how poverty is calculated in Europe or sweden specifically. The UK uses 60% of median income as it's poverty rate, so lets stick with that measure, because you'd think that would be consistent with the EU.
By that account one would need 38k (more or less) dollars per year to not be in 'relative' poverty. Before tax. Leaving 25k per year (admittedly, before tax) to be used to pay off debt. So it would take 60 years, but he's still be living on just under 40k equivalent his entirely life. That seems not great but manageable. Except we haven't taken into account wage growth over time.
Which takes us to to a question of 'what is his average salary going to be over 30 years'. That is a much harder question.
http://www.scb.se/Pages/TableAndChart____28313.aspx gives a longer more complex breakdown (that doesn't quite agree with wikipedia above but lets roll with it for the moment). In 1982 the 'average' salary in sweden was about 9000 Euro, compared to about 40k Euro now. Roughly a factor of 4 increase over 30 years. If you (wrongly but for sake of simple argument) assume that's linear. Then you can expect that in 15 years the median salary will be 126 dollars equivalent which would be the 'average' salary for the whole 30 year period.
So actually I undershot the projected income, but overshot on how much he'd keep. And I stand by my 30 year assessment then. It would take 15 years to be paying 50k (almost exactly, conveniently enough), but every year after that, assuming he only ever earned the exact average salary of an swede, he'd be paying more than 50k and have it paid of with 1 year before retirement at 65.
On top of that remember that's only his personal incomes. If he got married to someone also making the exact average income (and managed to stay together) they'd be doing pretty well for themselves.
So lets revisit what you said.
A normal person makes far less money than that. Also normal people do not get pensions. Those went away decades ago.
First part: True for today. Not true for the average swede over a 30 year period.
second part: not true for sweden at all. Not true for most decent places to live.
So given the choice between leeching off relatives until you die or living on the relative poverty line of 60% of median income (which is apparently about 38K USD right now) until retirement which would you prefer?
Obviously if he has to pay interest on that debt he's in trouble unless he can get a million dollar book deal. But a million and a half dollars is possible to pay off if you're in your 30's.
Again, I'm assuming he makes exactly the average income, if he manages anymore than that then it becomes significantly easier.
1.5 million you could pay back if you had a decent enough job. 100k a year minus 50k for 30 years. And you'd still collect pension benefits as though you had 100k.
That's manageable for a normal person. 1.5 billion... not so much.
Also, living with family is risky, because any assets you get in inheritance will still be claimed, and any needs you have as you get older will not be covered if your siblings/relatives can't, and you won't have any pension except government security.
oh and interestingly, note on that that UCLA has a "Nonresident Supplemental Tuition* 15,102.00" Which is I'm guessing the out of country/state fee but only applies to masters students.
Comp sci, physics and engineering. Oh and a neuroscientist. All have to pay tuition. And I can count about 30 in the US that I know personally, and we see quite a number (10 or 15 a year) here looking at grad school because they can't afford tuition in the US in a comp sci department.
Admittedly some of this is taking money from one pocket to pay another. I get paid in part based on the assumption that I have 7K in tuition, so that's budgeted as part of how much I get paid, but I still have to pay it. Foreign students however don't get paid differently than I do, and they have to pay 3x my tuition.
E.g. UCLA
http://www.registrar.ucla.edu/fees/gradfee.htm
Tuition 11,220.00
For grad students. Now a domestic (US) student would be paid enough that presumably they can cover some or all of that and still live.
Not that much more. A PhD in physics, chem or maths is mostly in the professor bracket. Which typically cap out in the 160k range as a full prof with 30 years experience. But you start at 50-70 and claw your way up from there. Engineering is a bit more than that, but not a lot. You can make more than that in industry of course, but the vast majority of PhD's in science are still academic type jobs.