You think it's slim? Let me give you some numbers then. If prime is 8.25 then you should be able to at least get a rate of 8.5% on a fully secured loan (secured with your 30k investments). At 8.5% over five years you will pay $6930 in total interest. Now lets assume that your investments do poorly and return only 8.5% (the same as your loan). 30K over five years at 8.5% turns into $45110. Subtract the $6930 and your still up $8180. I don't consider 8k slim.
You ignore taxes on that $15,110. Further, you are taking on market risk on debt that you may not be able to service. Get into a cash flow squeeze and - *boom!* - you're broke. Also, what happens if the market crashes? Suppose that collateral diminishes and the bank calls your loan? *Boom!* You're broke.
Hey, I'll borrow money to invest in an income generating asset. I did that with a two-fam five years ago. That's no different from a business borrowing to finance a plant retool on expected revenue or profit gains. But to borrow money in order to reinvest - well, it's your money. Good luck.
He doesn't buy used because he doesn't want to risk taking home a lemon. He doesn't think he has enough experience with automobile innards to make an informed choice on the used market. I've never bought a new car, however, for exactly that reason (not that I'm knowledgeable about cars either). But your post is spot on. These guys borrowing money in order to re-invest are in for one rude awakening some day...
Just so we're clear, a 5% loan for a car is lower than Fed Funds, which is currently 5.25%. Right now Prime is hovering around 8.25%. Thus, for a car financing entity to lend money at 5% must be a loss leader. These car financing and sales institutions make their money back somehow - where would that be? In the price of that car. Thus, one does not save money by taking out a larger (than necessary) loan with a perceived lower interest rate, regardless of whether one invests cash on hand that might have been used to buy a different (lower margin reflected in sticker price) car.
And then there is the issue of borrowing money in order to invest it elsewhere at a presumed profit. Multi-billion dollar hedge funds might be able to successfully pull off tricks like that and turn a sustained profit - but not you and me. You simply trade significant risk for a very slim potential profit.
Just to be clear: You are recommending that buyers should take out a loan in order to invest the money elsewhere at a profit, for example a five year term bank CD. Do I understand that correctly?
First off to the dealer cash, cheque or finance makes no difference to them.
It makes a huge difference if the dealer is financing the loan. They get all that interest then.
Secondly car dealers offer some insanely low interest rates (like 0%).
Around here that's only true for American made cars, and only specific models. Further, the lower interest rates are offset by higher car prices. They don't really give you free money!
But if you want to buy a decent imported car, there are no financing deals. Your best deal is to negotiate (haggle) and then pay upfront.
I know a guy who will walk onto a lot carrying a bank check in the amount he believes a car is worth (after conducting research), walk up to a car sitting on the lot, point to it, and waves the check in front of the salesman. He tries to time this for end of year sales, so he'll pick up the current model just as the lot is transitioning to next years model.
By doing this he saves interest on the car loan; he avoids buying extra junk to be installed, and drives the exact car he pointed to off the lot right then and there. Then he'll drive that car into the ground over the next decade or so until it's worthless. During that time he saves for the next car - just like he was paying for a car loan - except without the interest.
He's not rich, he doesn't make a pile of money - he just plans ahead. And it saves him a bundle of money.
At $3000, that's about five times the retail price and significantly more than my monthly mortgage payment. Supposing I did have $3000 in surplus cash, were I to use those funds to pay down principal on my mortgage instead of buying that PS3 I would shave a bit more than a years worth of mortgage payments from the end of the loan. Thus, to pay five times retail for a toy would incur a long term loss in excess of $24,000.
if the earth stops spinning all we have to do is pull the elevator cable really hard. It's like pushing the opposite wall of a train while it is in a steep turn, thus preventing it from tipping over. Duh!
...A $2000 laser pointer which outputs coherent light in the most difficult color spectrum for humans to see. I'll take one and a dozen pet rocks, please!
Slashdotters, trollers, and pollsters one and all, what say you? Disingenuous or dissembling?
All of the above. I have the right to use my media as I see fit. What, is he next going to claim that ripping my own CDs is illegal? Pesky "Fair Use", it's holding back commerce and hurting the economy. Hey, we have record executives to feed! They have children to put through ivy league schools! And Hummers to buy! The rabble is getting just too damn uppity, I say. Do you honestly think that's disingenuous?
Sheesh!
Now, if you'll excuse me, I have to go download last nights Frisky Dingo episode.
Yeah, I would guess that the producers weren't technical enough to really understand what that guy was doing. It's possible there was a prosecution angle to it too. It was a cool documentary though.
BTW: I loved that hahahahahaha post you did. Wuz wondering how you got it through the lameness filter. heh. So many angry responses to this thread, one could accuse me of inadvertent trolling. Actually, the mods did.:)
Then do you have information on the nature of these "statistical methods?"
Rules differ from state to state and are generally set and enforced by each Secretary's of State. For example, the state of Arizona has online the entire election procedures in pdf form. Check pages 112 and 113 for the rules dealing with ballot spoilage due to over and undervotes and other problems. These ballots are all tossed unless a manual recount is deemed necessary. As a result, election officials will perform a statistical analysis of race results to ascertain whether the electronic results are within out outside the known margin of error. If they are outside, then the race is called without any manual recount.
Different states set differing acceptable margins of error. For example, in the VA senate race George Allen lost by ~9000 votes. However, the margin of error used in VA is 2.5% of total ballots, which if the race is called within that margin, the losing candidate can then request a manual recount at state expense. I seem to remember press reports stating that 2.5% of the VA voting population ran ~12,000 votes, however, Allen conceded, and therefore the recount was not done. In Montana, where Burns lost to Talent, I believe the margin of error standard is set at.5%. So, even though Talent was ahead by ~1700 votes, because the margin of error was so much smaller, Talent was unable to request a manual recount at state expense.
Then do you know the federal cases where this standard came from?
No. But if you dig through this thread one poster was kind enough to link to a federal statute that defines these sorts of procedures. I haven't read his link so I can't speak to its veracity, but it looked good. Note that states control the mainstay of election procedures, however.
Not really. Your presumption appears to be that I brought up the soviet electoral system out of support. Nope, i do not. I was simply pointing out that anonymous voting does not necessarily confer citizen voting rights. IOW: Soviet elections were both anonymous and rigged.
Who said it was an acceptable error rate? Not a single commenter in this thread, including me, argues that this small an election represent enough sample to introduce margin of error. Again, there is a paper record. A county election official can recount this record and resolve the question.
Who said there was? Nobody is arguing that 36 votes are impossible to accurately count. There is a paper trail. So recount the paper trail and verify the vote totals. The point about margin of error had to do with a hundred millions votes counted across perhaps half a million voting machines.
Yes. That's called an undervote and is counted as such because you did not make an official selection for one of the races. Generally, if you opt to write in a candidate, what election officials will do is first set your ballot aside. Then they will count all the other votes. If, after tallying all the votes, the races are called outside the margin of error needed for any of your votes (that is, any vote within the races you voted), then they will set your ballot aside and certify the race.
Why should they manually count your ballot if they already know the outcome of the races? At least, that's the reasoning...
Actually, a hack like that was done by a Dutch security expert for HBO's Hacking Democracy. Unfortunately, was unable to find the gentleman's name on the website. However, he wrote a program that would sit on a smart card and execute upon insertion. This program then changed the machine vote count tallies, along with changing the recorded paper tally, and then erased itself. It was demonstrated on a standard Optical Scan Reader based system, one taken from a collection of certified machines sitting within a warehouse.
There is no doubt about feasibility of such a hack. The issue is scale of operation in this case, with a very small town and a local election at stake, vs. a hack that almost certainly would only be used (if ever - no evidence currently exists showing it has been) on a national scale.
I do not doubt this machine recorded in error. So, fine - determine the cause and fix it. There is a paper trail that - admittedly - could have been tampered with. In this case, so what?
Sure, but the accuracy of computers is far higher than you think. Your computer is calculating RAM addresses several million times every second. Get one such calculation wrong and the program crashes. Such crashes do happen, but it's extremely unusual.
Scale that up to something like a half million voting machines. Add the network (usually telephone modem) connection to one centralized tabulator (cluster) per state. Add in differing vote machine vendors, differing regional and state regulations, and anonymous voting.
It's not an easy solution. "Deterministic compute" deals with far more simple systems on a far smaller time scale. Run a large cluster or a single machine for hundreds of years and you'll see statistical noise within your compute results too.
Actually, Soviet voting was completely anonymous. Not that the results weren't predetermined. The solution being to only allow Communist party members access to candidacy.
Actually, not even finances are tracked with a perfect margin of error. Banks lose money! They really do.
Banks are not concerned with small statistically insignificant sums lost or found across all accounts. Further, the banking system is not concerned with this. Banks are very concerned about sums which move into a single (or range of) accounts. That is, banks and their regulators (and auditors) care about fraud.
There is always a statistical variance between what accountants believe is stored across all accounts within a bank (or the banking system) vs. what might be counted at any one time. Banks lose and gain small amounts of money all the time!
This is just the same as counting votes in a large-scale national voting system. However, as others have noted, voting is significantly more complex. Unlike banks (and ATMs), voters expect total anonymity. Which, unlike accessing your bank account, means that determining the intent of the voter (as opposed to the intent of a bank customer) after the fact is next to impossible due to the lack of records. Even in the perfect scenario of a paper trail for every vote. This is why there are under and over votes in every election. Votes lost when it is clear that a voter did vote.
The only viable solution is to verify that the aggregate count is accurate to within the margin of error. Otherwise, demand a recount.
Slashdot Mirror
You think it's slim? Let me give you some numbers then. If prime is 8.25 then you should be able to at least get a rate of 8.5% on a fully secured loan (secured with your 30k investments). At 8.5% over five years you will pay $6930 in total interest. Now lets assume that your investments do poorly and return only 8.5% (the same as your loan). 30K over five years at 8.5% turns into $45110. Subtract the $6930 and your still up $8180. I don't consider 8k slim.
You ignore taxes on that $15,110. Further, you are taking on market risk on debt that you may not be able to service. Get into a cash flow squeeze and - *boom!* - you're broke. Also, what happens if the market crashes? Suppose that collateral diminishes and the bank calls your loan? *Boom!* You're broke.
Hey, I'll borrow money to invest in an income generating asset. I did that with a two-fam five years ago. That's no different from a business borrowing to finance a plant retool on expected revenue or profit gains. But to borrow money in order to reinvest - well, it's your money. Good luck.
He doesn't buy used because he doesn't want to risk taking home a lemon. He doesn't think he has enough experience with automobile innards to make an informed choice on the used market. I've never bought a new car, however, for exactly that reason (not that I'm knowledgeable about cars either). But your post is spot on. These guys borrowing money in order to re-invest are in for one rude awakening some day...
Just so we're clear, a 5% loan for a car is lower than Fed Funds, which is currently 5.25%. Right now Prime is hovering around 8.25%. Thus, for a car financing entity to lend money at 5% must be a loss leader. These car financing and sales institutions make their money back somehow - where would that be? In the price of that car. Thus, one does not save money by taking out a larger (than necessary) loan with a perceived lower interest rate, regardless of whether one invests cash on hand that might have been used to buy a different (lower margin reflected in sticker price) car.
And then there is the issue of borrowing money in order to invest it elsewhere at a presumed profit. Multi-billion dollar hedge funds might be able to successfully pull off tricks like that and turn a sustained profit - but not you and me. You simply trade significant risk for a very slim potential profit.
Seems like a bad plan to me. JMO though.
Just to be clear: You are recommending that buyers should take out a loan in order to invest the money elsewhere at a profit, for example a five year term bank CD. Do I understand that correctly?
First off to the dealer cash, cheque or finance makes no difference to them.
It makes a huge difference if the dealer is financing the loan. They get all that interest then.
Secondly car dealers offer some insanely low interest rates (like 0%).
Around here that's only true for American made cars, and only specific models. Further, the lower interest rates are offset by higher car prices. They don't really give you free money!
But if you want to buy a decent imported car, there are no financing deals. Your best deal is to negotiate (haggle) and then pay upfront.
I know a guy who will walk onto a lot carrying a bank check in the amount he believes a car is worth (after conducting research), walk up to a car sitting on the lot, point to it, and waves the check in front of the salesman. He tries to time this for end of year sales, so he'll pick up the current model just as the lot is transitioning to next years model.
By doing this he saves interest on the car loan; he avoids buying extra junk to be installed, and drives the exact car he pointed to off the lot right then and there. Then he'll drive that car into the ground over the next decade or so until it's worthless. During that time he saves for the next car - just like he was paying for a car loan - except without the interest.
He's not rich, he doesn't make a pile of money - he just plans ahead. And it saves him a bundle of money.
At $3000, that's about five times the retail price and significantly more than my monthly mortgage payment. Supposing I did have $3000 in surplus cash, were I to use those funds to pay down principal on my mortgage instead of buying that PS3 I would shave a bit more than a years worth of mortgage payments from the end of the loan. Thus, to pay five times retail for a toy would incur a long term loss in excess of $24,000.
Think about that.
What's really funny is that I had planned to post that AC and instead clicked No Karma Bonus. D'OH!
HAHAHAHAHAHAHAHAHAHAHAHAHA!!!
Someone mark that ++,Funny!
HAHAHAHAHAHAHAHAHAHAHAHAHA!!!
Dude,
if the earth stops spinning all we have to do is pull the elevator cable really hard. It's like pushing the opposite wall of a train while it is in a steep turn, thus preventing it from tipping over. Duh!
...A $2000 laser pointer which outputs coherent light in the most difficult color spectrum for humans to see. I'll take one and a dozen pet rocks, please!
Slashdotters, trollers, and pollsters one and all, what say you? Disingenuous or dissembling?
All of the above. I have the right to use my media as I see fit. What, is he next going to claim that ripping my own CDs is illegal? Pesky "Fair Use", it's holding back commerce and hurting the economy. Hey, we have record executives to feed! They have children to put through ivy league schools! And Hummers to buy! The rabble is getting just too damn uppity, I say. Do you honestly think that's disingenuous?
Sheesh!
Now, if you'll excuse me, I have to go download last nights Frisky Dingo episode.
Yeah, I would guess that the producers weren't technical enough to really understand what that guy was doing. It's possible there was a prosecution angle to it too. It was a cool documentary though.
:)
BTW: I loved that hahahahahaha post you did. Wuz wondering how you got it through the lameness filter. heh. So many angry responses to this thread, one could accuse me of inadvertent trolling. Actually, the mods did.
Then do you have information on the nature of these "statistical methods?"
.5%. So, even though Talent was ahead by ~1700 votes, because the margin of error was so much smaller, Talent was unable to request a manual recount at state expense.
Rules differ from state to state and are generally set and enforced by each Secretary's of State. For example, the state of Arizona has online the entire election procedures in pdf form. Check pages 112 and 113 for the rules dealing with ballot spoilage due to over and undervotes and other problems. These ballots are all tossed unless a manual recount is deemed necessary. As a result, election officials will perform a statistical analysis of race results to ascertain whether the electronic results are within out outside the known margin of error. If they are outside, then the race is called without any manual recount.
Different states set differing acceptable margins of error. For example, in the VA senate race George Allen lost by ~9000 votes. However, the margin of error used in VA is 2.5% of total ballots, which if the race is called within that margin, the losing candidate can then request a manual recount at state expense. I seem to remember press reports stating that 2.5% of the VA voting population ran ~12,000 votes, however, Allen conceded, and therefore the recount was not done. In Montana, where Burns lost to Talent, I believe the margin of error standard is set at
Then do you know the federal cases where this standard came from?
No. But if you dig through this thread one poster was kind enough to link to a federal statute that defines these sorts of procedures. I haven't read his link so I can't speak to its veracity, but it looked good. Note that states control the mainstay of election procedures, however.
Not really. Your presumption appears to be that I brought up the soviet electoral system out of support. Nope, i do not. I was simply pointing out that anonymous voting does not necessarily confer citizen voting rights. IOW: Soviet elections were both anonymous and rigged.
*shrug*
Oh no. I the Soviet system everyone voted for the candidate. Anonymously.
That's interesting. Do you read a discrepancy between the section you quoted in that wikipedia article vs. what I actually wrote?
I see no such disparity.
Who said it was an acceptable error rate? Not a single commenter in this thread, including me, argues that this small an election represent enough sample to introduce margin of error. Again, there is a paper record. A county election official can recount this record and resolve the question.
Who said there was? Nobody is arguing that 36 votes are impossible to accurately count. There is a paper trail. So recount the paper trail and verify the vote totals. The point about margin of error had to do with a hundred millions votes counted across perhaps half a million voting machines.
Yes. That's called an undervote and is counted as such because you did not make an official selection for one of the races. Generally, if you opt to write in a candidate, what election officials will do is first set your ballot aside. Then they will count all the other votes. If, after tallying all the votes, the races are called outside the margin of error needed for any of your votes (that is, any vote within the races you voted), then they will set your ballot aside and certify the race.
Why should they manually count your ballot if they already know the outcome of the races? At least, that's the reasoning...
Actually, a hack like that was done by a Dutch security expert for HBO's Hacking Democracy. Unfortunately, was unable to find the gentleman's name on the website. However, he wrote a program that would sit on a smart card and execute upon insertion. This program then changed the machine vote count tallies, along with changing the recorded paper tally, and then erased itself. It was demonstrated on a standard Optical Scan Reader based system, one taken from a collection of certified machines sitting within a warehouse.
There is no doubt about feasibility of such a hack. The issue is scale of operation in this case, with a very small town and a local election at stake, vs. a hack that almost certainly would only be used (if ever - no evidence currently exists showing it has been) on a national scale.
I do not doubt this machine recorded in error. So, fine - determine the cause and fix it. There is a paper trail that - admittedly - could have been tampered with. In this case, so what?
Sure, but the accuracy of computers is far higher than you think. Your computer is calculating RAM addresses several million times every second. Get one such calculation wrong and the program crashes. Such crashes do happen, but it's extremely unusual.
Scale that up to something like a half million voting machines. Add the network (usually telephone modem) connection to one centralized tabulator (cluster) per state. Add in differing vote machine vendors, differing regional and state regulations, and anonymous voting.
It's not an easy solution. "Deterministic compute" deals with far more simple systems on a far smaller time scale. Run a large cluster or a single machine for hundreds of years and you'll see statistical noise within your compute results too.
thank you
Actually, Soviet voting was completely anonymous. Not that the results weren't predetermined. The solution being to only allow Communist party members access to candidacy.
See: Soviet Democracy
"Deputies to all Soviets shall be elected on the basis of universal, equal, and direct suffrage by secret ballot."
Actually, not even finances are tracked with a perfect margin of error. Banks lose money! They really do.
Banks are not concerned with small statistically insignificant sums lost or found across all accounts. Further, the banking system is not concerned with this. Banks are very concerned about sums which move into a single (or range of) accounts. That is, banks and their regulators (and auditors) care about fraud.
There is always a statistical variance between what accountants believe is stored across all accounts within a bank (or the banking system) vs. what might be counted at any one time. Banks lose and gain small amounts of money all the time!
This is just the same as counting votes in a large-scale national voting system. However, as others have noted, voting is significantly more complex. Unlike banks (and ATMs), voters expect total anonymity. Which, unlike accessing your bank account, means that determining the intent of the voter (as opposed to the intent of a bank customer) after the fact is next to impossible due to the lack of records. Even in the perfect scenario of a paper trail for every vote. This is why there are under and over votes in every election. Votes lost when it is clear that a voter did vote.
The only viable solution is to verify that the aggregate count is accurate to within the margin of error. Otherwise, demand a recount.
*shrug*