Man, do I wish I had mod points today. You make a very interesting point compared to the random predictable noise in the rest of this comment thread...
This is quite ironic, because neural networks are essentially about learning functions between different spaces, so they actually do have calculus behind them. It just may be sufficiently hidden that you didn't notice.
There are quite a lot of places in CS theory where calculus or, to be more precise, a solid understanding of the exponential function and related topics, is very valuable. I remember taking a course on networking at university, and it happened to talk about probabilistic algorithms, queueing theory, etc., where people suddenly had to deal with exponential distributions and such things, which are really built on calculus. Of course, teaching calculus in the first year probably isn't going to help that much when people actually only need it two or three years later, but that's a general problem in education.
Looking at the political discourse in Western democracies these days, a more accurate statement would be that democracy is 3 wolves persuading 1000 sheep that the wolves should get meat for dinner. Then, when some of the sheep point out that that's unfair, the other sheep will attack them, having been thoroughly brainwashed by the wolves.
Their algorithm for the Novice player basically looks at your last four throws, and then looks at the history of how often those last four throws were followed by each of rock, paper, scissors, and picks the corresponding counter. It is trivial to fool this, with the following Python snippet (note that the Python code will just pick randomly when there is no clear winner):
import random freq = [0] * 3*3*3*3*3 history = 0 while True: raw_input('') history = history * 3 h = freq[history:history+3] print h choice = random.choice([(j+2)%3 for j in range(3) if h[j] == max(h)]) print 'Throw', ['rock', 'paper', 'scissors'][choice] history = history + choice freq[history] += 1 history = history % 81
The Veteran computer player uses a different strategy, based on some database. Once you've inferred the database from its responses, beating it should be trivial, but of course it would take relatively long to read out the entire database.
I'm not sure what you mean by "natural rate of unemployment". I am sure that you can define that term in such a way that what you're saying is true, but I would argue that such a definition is ultimately not useful at all. As far as the NAIRU is concerned, the concept is hugely flawed, and the perceived increase of that rate can be mostly traced to economists changing their models as it pleases them when reality contradicts their claims.
And who is going to pay for your scientists and engineers to dream things up while other countries are busy working and building the next new things?
The government can pay for that. See, for example, Debt is not debt to get an alternative view to what all the deficit terrorists are telling you these days.
I don't know what the solution is, but I assume it involves either a sudden collective burst of altruism in employers (ho ho) or some truly massive government intervention (hee hee).
In all seriousness, government intervention is the way to go in this case. Either by regulation (lower the retirement age, force shorter working hours / 4-day week), or by sound economic policies. In a way, it is in fact the government's responsibility to do that. Mass unemployment is the result of demand shortage, which is caused by a preference for financial assets in the private sector. Either the government has to eliminate that preference for financial assets, which is probably only possible by decreasing wealth inequality, or it has to step in and fill the demand gap itself.
The small government folk don't like to hear it, but the truth is that individuals can do nothing against macroeconomic problems. Fixing macroeconomic problems is exactly the kind of thing that even small governments need to take care of.
I hope you are praising the unemployed of Denmark as the heroes who stabilize the economy... at least, that is what you would logically have to do if you take that kind of argument seriously.
I'm sorry to burst your bubble, but there really are some (in my view very misguided and/or sociopathic) individuals who believe that it is not government's job to work towards full employment - meaning that the only type of unemployment is the kind of frictional/between jobs unemployment that you mention. This rate of unemployment is probably more like 0.5%-2%, at least judging from the period of low unemployment in many Western countries after World War II (Germany, for example, had a run of less than 1% unemployment for several years in a row in the 1960s). This is not to be confused with the so called NAIRU, a fictional (that is, unmeasurable) unemployment rate that is supposedly neutral with respect to inflation. The latter is also a somewhat misguided concept, because there are ways of creating jobs without creating any inflationary pressure.
Here is an interesting article by Victor Quirk on the historical development of policies related to full employment, and the abandonment of those policies with the policy shifts starting in the 1970s.
There are a number of people who really do think like this, but it is a totally unnecessary waste of resources. First of all, 10 percent is way too large. The way unemployment is measured these days, it's likely that even an unemployment rate of 3% wouldn't create undue price pressures. Still, there are ways of creating jobs without any price pressure at all, via a Job Guarantee / Employer of Last Resort scheme. See for example the writings of Randall Wray and Bill Mitchell for some ideas on that.
Keep in mind that it is up to the stock exchange to set the smallest increment in prices, which - if the market has many competing players - is probably the largest influence on the spread. This minimum spread happens to have been decreased in the US by the SEC. This probably had a larger impact on reducing spreads than anything else.
I'm incredibly curious to hear why you feel that adding more opaqueness to the system will lead to less price volatility and more efficiency.
How does what I have outlined add more opaqueness to the system? The truth is that when you participate in any kind of trading system, you are always in a race against orders that you don't know, because communication is not instantaneous. When you place an order, you have not seen at least the last ping time worth of orders when you make your decision, and it takes some time for your order to get to the trading system and be processed. That time may be small, but in a busy trading system, this means that your order will always race against other orders that you simply cannot know about. This cannot be avoided as long as players in the system are allowed to place orders in parallel, and the reason that brokers want their automatic trading systems as close to the stock exchange as possible is that it allows them to beat others in that race.
There are two things my proposal changes: One is to make the fact that your orders race against other orders/other bids which you cannot know very explicit and obvious - so it actually makes the system more transparent (I understand why this seems counter-intuitive at first, but a lot of stuff that is motivated by mechanism design and game theory is that way until you take some time to really think it through - second price auctions are perhaps the most famous example of such an unintuitive-at-first concept). The second is that it levels the playing field, at least as far as technological access is concerned: Millisecond differences in ping time will no longer be relevant, because the heartbeat frequency is long enough that they can be ignored.
Of course there are other artificial barriers to entry that would need to be removed to get a truly free market, such as exorbitant up-front fees that need to be paid to participate in the system in the first place. I am merely trying to address one factor that makes existing trading systems different from the ideal markets that the economists dream about.
Perhaps I haven't been very clear in explaining the proposal because I thought it is obvious that of course all offers that are valid for the next heartbeat are hidden to other players, and only become visible after the heartbeat. Think of it like a multi-player auction (with multiple sellers and buyers) that repeats once per minute: all bids are sealed until the auction-resolving algorithm kicks in. The important part to understand is that this is really not qualitatively different from the existing system, but it makes the functioning of the system more transparent and levels the playing field. Both properties are Good Things in my good.
If I cant post my shares in that 1 minute interval because I too must wait 1 minute, then it doesnt mean anything that there was no volatility in that minute. I can't get the price on the board, but must guess at the next price. If I can post my shares in that 1 minute interval, then there really isnt a 1 minute interval after all because there are multiple players.
The point is that you can never get the price on the board, and must always guess at the next price. Higher frequency that means that the jumps occur more often, and that the playing field becomes less level.
The upshot of this is that if there are outstanding trades still in progress when I decide to make my own trade, then there is volatility. This rapid arbitrage causes the majority of all outstanding trades near the margin to be filled nearly instantly, so the price on the ticker more closely resembles the price that I can pay or get. That is reduced volatility. The only effect this rapid arbitrage has is that instead of a hundred thousand people arbitraging, its only a few big traders located next door to the exchange... and I don't care if its 1 player or 100000 on the other end.. what I care about is that the price I see resembles the price I can get.
If there is only one other player at the end, then they can set the spread between the price at which you can sell vs. the price at which you can buy to whatever they like. The more players there are, the higher the pressure for this spread to be reduced, so it is important to level the playing field and get more players. The frequency at which those players trade is really of lesser importance.
Market makers and providing liquidity is important. However, high-frequency trading simply isn't necessary for that. To be more precise, the problem with your line of reasoning is that if you can change your prices only once per minute, then there will be no volatility during that minute, and hence there is no need to change the prices. So your argument regarding spreads is invalid.
The whole millisecond trading thing is just a way to raise the barrier of entry to the market: The people who are big enough that they can afford close proximity and fast connections to the trading system benefit, everybody else is left out. So competition is reduced, which makes the market less efficient.
What should happen is that all those trading systems operate on a heartbeat with a fixed frequency of e.g. one minute. Basically, everybody gets to make their offers, and once per minute all those offers are matched against each other and the outcome, i.e. the transactions that take place as a result, are computed using one of the standard auction mechanisms. After receiving the outcome, traders then have time until the next heartbeat to adjust their offers. This way, the insane barriers of entry are removed, and the functionality of the system remains essentially the same.
Actually, more debt means more deposits. When you take a loan, you get money from your bank, which you then use to pay somebody. This somebody puts the money on an account at a bank - voila, more deposits.
However, a good savings rate more help the economy more long term so I do not know.
A high savings rate by itself doesn't help the economy - that is unfortunately a common misconception.
If you simply keep your money in a bank account, that doesn't do anybody much good - it can help the banks make a larger profit, but it will not increase the amount of money that banks lend. In a depressed economy, the lending of banks is constrained by the number of creditworthy borrowers. In a lively economy, the lending of banks is constrained by their capital and regulatory capital requirements. From the bank's point of view, money in bank accounts is a liability that does not count towards capital.
Buying new emissions of shares or other direct forms of actual investments are a different story, of course.
Are you assuming that the write-up of those decision will be honest? Because when it comes to such things, honesty really comes first. Requiring massive pages of text by itself is worthless, if you cannot be reasonably confident that they are written with honesty.
which is why the current political obsession with austerity is so ridiculous
How do you figure? Over time, there's a limit on the amount of money that a government can raise though taxes (thanks to the Laffer curve having a maximum), and a limit to the amount of money a government with rolling debt can print (because it has to continue to borrow against the perceived risk of future inflation). This means the total amount of money a government can spend (without a currency collapse) is limited, not infinite.
I hate to write it this plainly, but from a technical point of view, this is simply false. The government (as a whole, i.e. assuming both executive and legislative bodies get behind it) can always spend as much as it likes. As long as there are goods and services for sale denominated in US$, the US government can buy them. Think about it:
You offer something for sale at a certain US$ value. The government writes you a check to that amount. You go to a bank to cash that check. If the check bounces, then the government is unable to pay you. If the check doesn't bounce, then the government is able to pay you.
Since it's ultimately the government (via the Fed) that runs the US$ monetary system, government checks never bounce. This means that on a functional level, the US government can spend as much as it likes.
Is spending as much as it likes always a good idea? Obviously not. But once you accept the fact that government spending has no technical limitations - unlike the spending of you and I - you are free to ask other questions about how the economy should be run. If the government has no spending limit, why are there taxes? There is a good reason for them. Why does it borrow? There are fewer good reasons for that. And how much should it tax? More than it spends? Less than it spends? Exactly the same amount? Figuring out an answer to this question without any ideological baggage is precisely what Modern Monetary Theory is about, see e.g. here.
The most important lesson is perhaps this: Mindless budget rules are stupid. The appropriate size of the budget and its balance can only be determined in the context of how the real economy is doing, the amount of unemployment, and political choices.
Absolutely. This is the one thing that "goldies" never seem to get right. Money is all about you owing me and vice versa. Moreover, all money in existence is ultimately a debt of the government, which is why the current political obsession with austerity is so ridiculous. Government debt is simply the mirror image of private wealth.
It's true that paper money can be seen as a form of government debt, but the amount of money in an economy is much, much higher than the amount of currency. The amount of money is determined by the banks' ability and willingness to lend out money - the same coins and bills can be shovelled back and forth between the bank and the loaners/savers multiple times, creating debts and savings far larger than the nominal values of the coins/bills.
True. I should have said that all net money/financial assets in existence is a debt of the government. There is a difference between horizontal transactions, which do not affect the net sum of financial assets in the private sector, and vertical transaction, i.e. the government creating financial assets by spending, and destroying them by taxation. Note that government "borrowing" does not change the net sum of financial assets either, it just changes their composition (towards bonds).
Wealth, in turn, is not directly related to the amount of money - wealth is determined by the amount of physical goods and services produced in an economy. Money is merely a means to circulate these goods and services.
I think we can agree that wealth can be defined in different ways and is a rather vague concept anyway. And yes, money is ultimately "just" the way by which we distribute goods and services in our societies, but money is not neutral. In particular, money is used to distribute goods and services across time (something that wouldn't be possible in e.g. a pure barter or gift economy), as people use money as a storage of wealth. In other words, money is far from neutral, and you have to consider its effects on the economy if you want to understand the formation of wealth.
Let's say BitCoin were to take off in a serious way and have, some years in the future, one million users. So there will be, on average, 21 BTC per user. Now assume it really does go global and there will be one billion users. Suddenly, you only have 0.021 BTC per user. Do you honestly not see how that would cause deflationary pressure?
Furthermore, even if the population itself were constant, the same effects would appear in a growing economy. You seem to assume that the total supply and/or demand in an economy is a constant in the long run. I hate to break it to you, but reality looks very different.
Absolutely. This is the one thing that "goldies" never seem to get right. Money is all about you owing me and vice versa. Moreover, all money in existence is ultimately a debt of the government, which is why the current political obsession with austerity is so ridiculous. Government debt is simply the mirror image of private wealth.
FIAT currency tends to be *backed* by something, like an economy, like USA or European Union or even China.
More concretely, modern fiat money is backed by the power of taxation. A large part of the value of money comes from ultimately circular reasoning, i.e. you can pay your groceries at the shop using money, because the shop needs to pay its employees with that money, because those employees can use that money to pay for groceries. However, underlying this circularity is the fact that the state has a monopoly power to force a debt on you: the tax debt. A random stranger on the street cannot force you into debt, but the state can. By doing so, it creates scarcity and demand for the state-issued money, and this is where the value of modern money ultimately comes from.
This is one of the key insights of Modern Monetary Theory, which was greatly influenced by the Functional Finance of Abba Lerner, and is developed by people like Randall Wray and Bill Mitchell.
The US does not have this problem, because the executive is elected directly by the people.
I'm not familiar with the Israeli political system, but many European countries have this problem because the people only vote for the parliament, and then the parliament votes for the prime minister / chancellor. In order to successfully form a government, there consequently has to be some majority in the parliament, which can be difficult to form given many small parties.
The US system is fundamentally different. Having executive and legislative aligned differently is relatively common anyway, and the president, and therefore the executive, is elected directly by the people. So whoever happens to be elected president might have to work on an individual basis to gather support for his legislative initiatives if there are many smaller parties, but this is already the case anyway in the contemporary system. In other words, having more, smaller parties wouldn't actually change that much.
Slashdot Mirror
Man, do I wish I had mod points today. You make a very interesting point compared to the random predictable noise in the rest of this comment thread...
This is quite ironic, because neural networks are essentially about learning functions between different spaces, so they actually do have calculus behind them. It just may be sufficiently hidden that you didn't notice.
There are quite a lot of places in CS theory where calculus or, to be more precise, a solid understanding of the exponential function and related topics, is very valuable. I remember taking a course on networking at university, and it happened to talk about probabilistic algorithms, queueing theory, etc., where people suddenly had to deal with exponential distributions and such things, which are really built on calculus. Of course, teaching calculus in the first year probably isn't going to help that much when people actually only need it two or three years later, but that's a general problem in education.
What about the truth?
I have to admit that that's a pretty harsh thing to tell a three year old, but what good is a lie going to do?
Looking at the political discourse in Western democracies these days, a more accurate statement would be that democracy is 3 wolves persuading 1000 sheep that the wolves should get meat for dinner. Then, when some of the sheep point out that that's unfair, the other sheep will attack them, having been thoroughly brainwashed by the wolves.
Their algorithm for the Novice player basically looks at your last four throws, and then looks at the history of how often those last four throws were followed by each of rock, paper, scissors, and picks the corresponding counter. It is trivial to fool this, with the following Python snippet (note that the Python code will just pick randomly when there is no clear winner):
import random
freq = [0] * 3*3*3*3*3
history = 0
while True:
raw_input('')
history = history * 3
h = freq[history:history+3]
print h
choice = random.choice([(j+2)%3 for j in range(3) if h[j] == max(h)])
print 'Throw', ['rock', 'paper', 'scissors'][choice]
history = history + choice
freq[history] += 1
history = history % 81
The Veteran computer player uses a different strategy, based on some database. Once you've inferred the database from its responses, beating it should be trivial, but of course it would take relatively long to read out the entire database.
I'm not sure what you mean by "natural rate of unemployment". I am sure that you can define that term in such a way that what you're saying is true, but I would argue that such a definition is ultimately not useful at all. As far as the NAIRU is concerned, the concept is hugely flawed, and the perceived increase of that rate can be mostly traced to economists changing their models as it pleases them when reality contradicts their claims.
And who is going to pay for your scientists and engineers to dream things up while other countries are busy working and building the next new things?
The government can pay for that. See, for example, Debt is not debt to get an alternative view to what all the deficit terrorists are telling you these days.
I don't know what the solution is, but I assume it involves either a sudden collective burst of altruism in employers (ho ho) or some truly massive government intervention (hee hee).
In all seriousness, government intervention is the way to go in this case. Either by regulation (lower the retirement age, force shorter working hours / 4-day week), or by sound economic policies. In a way, it is in fact the government's responsibility to do that. Mass unemployment is the result of demand shortage, which is caused by a preference for financial assets in the private sector. Either the government has to eliminate that preference for financial assets, which is probably only possible by decreasing wealth inequality, or it has to step in and fill the demand gap itself.
The small government folk don't like to hear it, but the truth is that individuals can do nothing against macroeconomic problems. Fixing macroeconomic problems is exactly the kind of thing that even small governments need to take care of.
I hope you are praising the unemployed of Denmark as the heroes who stabilize the economy... at least, that is what you would logically have to do if you take that kind of argument seriously.
I'm sorry to burst your bubble, but there really are some (in my view very misguided and/or sociopathic) individuals who believe that it is not government's job to work towards full employment - meaning that the only type of unemployment is the kind of frictional/between jobs unemployment that you mention. This rate of unemployment is probably more like 0.5%-2%, at least judging from the period of low unemployment in many Western countries after World War II (Germany, for example, had a run of less than 1% unemployment for several years in a row in the 1960s). This is not to be confused with the so called NAIRU, a fictional (that is, unmeasurable) unemployment rate that is supposedly neutral with respect to inflation. The latter is also a somewhat misguided concept, because there are ways of creating jobs without creating any inflationary pressure.
Here is an interesting article by Victor Quirk on the historical development of policies related to full employment, and the abandonment of those policies with the policy shifts starting in the 1970s.
There are a number of people who really do think like this, but it is a totally unnecessary waste of resources. First of all, 10 percent is way too large. The way unemployment is measured these days, it's likely that even an unemployment rate of 3% wouldn't create undue price pressures. Still, there are ways of creating jobs without any price pressure at all, via a Job Guarantee / Employer of Last Resort scheme. See for example the writings of Randall Wray and Bill Mitchell for some ideas on that.
Keep in mind that it is up to the stock exchange to set the smallest increment in prices, which - if the market has many competing players - is probably the largest influence on the spread. This minimum spread happens to have been decreased in the US by the SEC. This probably had a larger impact on reducing spreads than anything else.
I'm incredibly curious to hear why you feel that adding more opaqueness to the system will lead to less price volatility and more efficiency.
How does what I have outlined add more opaqueness to the system? The truth is that when you participate in any kind of trading system, you are always in a race against orders that you don't know, because communication is not instantaneous. When you place an order, you have not seen at least the last ping time worth of orders when you make your decision, and it takes some time for your order to get to the trading system and be processed. That time may be small, but in a busy trading system, this means that your order will always race against other orders that you simply cannot know about. This cannot be avoided as long as players in the system are allowed to place orders in parallel, and the reason that brokers want their automatic trading systems as close to the stock exchange as possible is that it allows them to beat others in that race.
There are two things my proposal changes: One is to make the fact that your orders race against other orders/other bids which you cannot know very explicit and obvious - so it actually makes the system more transparent (I understand why this seems counter-intuitive at first, but a lot of stuff that is motivated by mechanism design and game theory is that way until you take some time to really think it through - second price auctions are perhaps the most famous example of such an unintuitive-at-first concept). The second is that it levels the playing field, at least as far as technological access is concerned: Millisecond differences in ping time will no longer be relevant, because the heartbeat frequency is long enough that they can be ignored.
Of course there are other artificial barriers to entry that would need to be removed to get a truly free market, such as exorbitant up-front fees that need to be paid to participate in the system in the first place. I am merely trying to address one factor that makes existing trading systems different from the ideal markets that the economists dream about.
Perhaps I haven't been very clear in explaining the proposal because I thought it is obvious that of course all offers that are valid for the next heartbeat are hidden to other players, and only become visible after the heartbeat. Think of it like a multi-player auction (with multiple sellers and buyers) that repeats once per minute: all bids are sealed until the auction-resolving algorithm kicks in. The important part to understand is that this is really not qualitatively different from the existing system, but it makes the functioning of the system more transparent and levels the playing field. Both properties are Good Things in my good.
If I cant post my shares in that 1 minute interval because I too must wait 1 minute, then it doesnt mean anything that there was no volatility in that minute. I can't get the price on the board, but must guess at the next price. If I can post my shares in that 1 minute interval, then there really isnt a 1 minute interval after all because there are multiple players.
The point is that you can never get the price on the board, and must always guess at the next price. Higher frequency that means that the jumps occur more often, and that the playing field becomes less level.
The upshot of this is that if there are outstanding trades still in progress when I decide to make my own trade, then there is volatility. This rapid arbitrage causes the majority of all outstanding trades near the margin to be filled nearly instantly, so the price on the ticker more closely resembles the price that I can pay or get. That is reduced volatility. The only effect this rapid arbitrage has is that instead of a hundred thousand people arbitraging, its only a few big traders located next door to the exchange... and I don't care if its 1 player or 100000 on the other end.. what I care about is that the price I see resembles the price I can get.
If there is only one other player at the end, then they can set the spread between the price at which you can sell vs. the price at which you can buy to whatever they like. The more players there are, the higher the pressure for this spread to be reduced, so it is important to level the playing field and get more players. The frequency at which those players trade is really of lesser importance.
Market makers and providing liquidity is important. However, high-frequency trading simply isn't necessary for that. To be more precise, the problem with your line of reasoning is that if you can change your prices only once per minute, then there will be no volatility during that minute, and hence there is no need to change the prices. So your argument regarding spreads is invalid.
The whole millisecond trading thing is just a way to raise the barrier of entry to the market: The people who are big enough that they can afford close proximity and fast connections to the trading system benefit, everybody else is left out. So competition is reduced, which makes the market less efficient.
What should happen is that all those trading systems operate on a heartbeat with a fixed frequency of e.g. one minute. Basically, everybody gets to make their offers, and once per minute all those offers are matched against each other and the outcome, i.e. the transactions that take place as a result, are computed using one of the standard auction mechanisms. After receiving the outcome, traders then have time until the next heartbeat to adjust their offers. This way, the insane barriers of entry are removed, and the functionality of the system remains essentially the same.
Actually, more debt means more deposits. When you take a loan, you get money from your bank, which you then use to pay somebody. This somebody puts the money on an account at a bank - voila, more deposits.
However, a good savings rate more help the economy more long term so I do not know.
A high savings rate by itself doesn't help the economy - that is unfortunately a common misconception.
If you simply keep your money in a bank account, that doesn't do anybody much good - it can help the banks make a larger profit, but it will not increase the amount of money that banks lend. In a depressed economy, the lending of banks is constrained by the number of creditworthy borrowers. In a lively economy, the lending of banks is constrained by their capital and regulatory capital requirements. From the bank's point of view, money in bank accounts is a liability that does not count towards capital.
Buying new emissions of shares or other direct forms of actual investments are a different story, of course.
Are you assuming that the write-up of those decision will be honest? Because when it comes to such things, honesty really comes first. Requiring massive pages of text by itself is worthless, if you cannot be reasonably confident that they are written with honesty.
which is why the current political obsession with austerity is so ridiculous
How do you figure? Over time, there's a limit on the amount of money that a government can raise though taxes (thanks to the Laffer curve having a maximum), and a limit to the amount of money a government with rolling debt can print (because it has to continue to borrow against the perceived risk of future inflation). This means the total amount of money a government can spend (without a currency collapse) is limited, not infinite.
I hate to write it this plainly, but from a technical point of view, this is simply false. The government (as a whole, i.e. assuming both executive and legislative bodies get behind it) can always spend as much as it likes. As long as there are goods and services for sale denominated in US$, the US government can buy them. Think about it:
You offer something for sale at a certain US$ value. The government writes you a check to that amount. You go to a bank to cash that check. If the check bounces, then the government is unable to pay you. If the check doesn't bounce, then the government is able to pay you.
Since it's ultimately the government (via the Fed) that runs the US$ monetary system, government checks never bounce. This means that on a functional level, the US government can spend as much as it likes.
Is spending as much as it likes always a good idea? Obviously not. But once you accept the fact that government spending has no technical limitations - unlike the spending of you and I - you are free to ask other questions about how the economy should be run. If the government has no spending limit, why are there taxes? There is a good reason for them. Why does it borrow? There are fewer good reasons for that. And how much should it tax? More than it spends? Less than it spends? Exactly the same amount? Figuring out an answer to this question without any ideological baggage is precisely what Modern Monetary Theory is about, see e.g. here.
The most important lesson is perhaps this: Mindless budget rules are stupid. The appropriate size of the budget and its balance can only be determined in the context of how the real economy is doing, the amount of unemployment, and political choices.
Absolutely. This is the one thing that "goldies" never seem to get right. Money is all about you owing me and vice versa. Moreover, all money in existence is ultimately a debt of the government, which is why the current political obsession with austerity is so ridiculous. Government debt is simply the mirror image of private wealth.
It's true that paper money can be seen as a form of government debt, but the amount of money in an economy is much, much higher than the amount of currency. The amount of money is determined by the banks' ability and willingness to lend out money - the same coins and bills can be shovelled back and forth between the bank and the loaners/savers multiple times, creating debts and savings far larger than the nominal values of the coins/bills.
True. I should have said that all net money/financial assets in existence is a debt of the government. There is a difference between horizontal transactions, which do not affect the net sum of financial assets in the private sector, and vertical transaction, i.e. the government creating financial assets by spending, and destroying them by taxation. Note that government "borrowing" does not change the net sum of financial assets either, it just changes their composition (towards bonds).
Wealth, in turn, is not directly related to the amount of money - wealth is determined by the amount of physical goods and services produced in an economy. Money is merely a means to circulate these goods and services.
I think we can agree that wealth can be defined in different ways and is a rather vague concept anyway. And yes, money is ultimately "just" the way by which we distribute goods and services in our societies, but money is not neutral. In particular, money is used to distribute goods and services across time (something that wouldn't be possible in e.g. a pure barter or gift economy), as people use money as a storage of wealth. In other words, money is far from neutral, and you have to consider its effects on the economy if you want to understand the formation of wealth.
Let's say BitCoin were to take off in a serious way and have, some years in the future, one million users. So there will be, on average, 21 BTC per user. Now assume it really does go global and there will be one billion users. Suddenly, you only have 0.021 BTC per user. Do you honestly not see how that would cause deflationary pressure?
Furthermore, even if the population itself were constant, the same effects would appear in a growing economy. You seem to assume that the total supply and/or demand in an economy is a constant in the long run. I hate to break it to you, but reality looks very different.
Currency is IOU notes that devalue over time.
Absolutely. This is the one thing that "goldies" never seem to get right. Money is all about you owing me and vice versa. Moreover, all money in existence is ultimately a debt of the government, which is why the current political obsession with austerity is so ridiculous. Government debt is simply the mirror image of private wealth.
FIAT currency tends to be *backed* by something, like an economy, like USA or European Union or even China.
More concretely, modern fiat money is backed by the power of taxation. A large part of the value of money comes from ultimately circular reasoning, i.e. you can pay your groceries at the shop using money, because the shop needs to pay its employees with that money, because those employees can use that money to pay for groceries. However, underlying this circularity is the fact that the state has a monopoly power to force a debt on you: the tax debt. A random stranger on the street cannot force you into debt, but the state can. By doing so, it creates scarcity and demand for the state-issued money, and this is where the value of modern money ultimately comes from.
This is one of the key insights of Modern Monetary Theory, which was greatly influenced by the Functional Finance of Abba Lerner, and is developed by people like Randall Wray and Bill Mitchell.
The US does not have this problem, because the executive is elected directly by the people.
I'm not familiar with the Israeli political system, but many European countries have this problem because the people only vote for the parliament, and then the parliament votes for the prime minister / chancellor. In order to successfully form a government, there consequently has to be some majority in the parliament, which can be difficult to form given many small parties.
The US system is fundamentally different. Having executive and legislative aligned differently is relatively common anyway, and the president, and therefore the executive, is elected directly by the people. So whoever happens to be elected president might have to work on an individual basis to gather support for his legislative initiatives if there are many smaller parties, but this is already the case anyway in the contemporary system. In other words, having more, smaller parties wouldn't actually change that much.