Money doesn’t pour into markets
One often reads people talking about how investors poured lots of money into stocks, bonds, houses, gold, or some other asset. Sometimes this fact is used to explain a big run-up in asset prices. “The money had to go somewhere, didn’t it.” This view is wrong, but it’s wrong in interesting ways, which reveal a lot about how people think about macroeconomics:
1. First of all, at the most basic level it’s obviously wrong to talk about money literally “going into” markets. If you visit Wall Street as a tourist and ask to see the big pot of money that people have invested in the stock market, they’ll give you a quizzical look. Money goes through markets. (Especially money created by the Fed; the monetary base.)
2. Most sensible people who talk about money pouring into markets don’t mean it literally. But even if the phrase is used figuratively, it is still wrong. Indeed, let’s think about what people might mean by money pouring into a market. Perhaps they mean that people are making lots of purchases. Money is a medium of exchange, so lots of money is being used to facilitate stock purchases. And it is true that on days when stock prices rise by an unusual amount, transactions volume of often higher than normal. Unfortunately, that’s even more true of stock price crashes, which are often associated by an enormous volume of trading. The famous stock market crash of 1987 was associated with record-breaking volume, but I don’t recall people saying investors poured money into the stock market that day.
3. I have the same problem with people who talk about money being “saved.” At the individual level that’s true, but my decision to hold on to more dollars means someone else is holding on to less dollars. If we (collectively) increase our real holdings of base money, that is contractionary for NGDP. But suppose we save in some other way. If I buy a financial asset, someone else sells me the asset. Suppose I buy a used house. I pull $200,000 out of the bank (dissaving) and buy a house worth $200,000 (saving). The seller gives up a $200,000 house (dissaving), and puts the $200,000 received by selling the house in the bank (saving.) It’s never going to net out to any aggregate increase in saving unless new capital is built. Now do the same example with a new house. I pull $200,000 out and buy a new house. Same as before. The man who builds the house gets $200,000, so he actually increases his new saving by $200,000. Aggregate saving in the economy goes up by $200,000. What’s different is that the new house didn’t exist before being built, hence total wealth rises by $200,000. Following the money is completely unhelpful for thinking about national saving, all that matters is whether new capital goods have been produced.
4. Others point to the money created by the Fed, often called the monetary base. Prior to mid-2008, this was composed of non-interest-bearing cash plus n0n-interest-bearing reserves. Some argue that an easy money policy in the 2000s tended to boost real estate prices. But why would this be true? One possibility is that the Fed put lots of reserves into the banking system, and these reserves were somehow “lent out” for new home purchases. In fact, roughly 95% of all new base money created in the years leading up to until mid-2008 was currency that went into people’s wallets, not new bank reserves. This is even true in the short run. Monthly increases in the base are overwhelmingly currency, with just a little bit of extra bank reserves (until late-2008.) So it’s hard to see how this new “money” could have been pouring into the housing markets. Money doesn’t have to “go somewhere,” 95% of it stays in wallets (prior to 2008).
5. Others argue that the Fed held interest rates to very low levels. But how did they do this? Interest rates are set by the markets, not the Fed. They can influence interest rates by injecting money into the economy, but if money was very easy shouldn’t we have observed a very rapid growth in NGDP? Or at least a big increase in bank reserves, if you prefer that metric? How do we know that rates weren’t low simply because the supply of credit schedule shifted right faster than the demand for credit schedule? Why assume monetary policy is involved at all?
6. Prior to 2008, the amount of new money created by the Fed is typically trivial, compared to the size of the economy. And it was usually injected by buying T-securities, which is a huge and liquid market. Thus it’s hard for me to see how the so-called Cantillon effects were important. On a typical day the Chinese would buy more Treasury debt than the Fed. If the Fed had bought German and Japanese debt instead of T-securities, then the holders of those securities would switch over and buy the T-securities the Fed was no longer buying. The net effects would be tiny.
7. Monetary policy can have a big effect on asset prices, but the effect has little to do with what assets the Fed is buying (until 2008, after that the purchases were so massive it might have marginally shifted interest rate spreads.) My hunch is that if the Fed had bought index stock funds during 2002-06 instead of T-securities, the impact on asset prices would have been roughly the same. If there was an impact, it was because monetary injections can raise expected future NGDP, and higher expected future NGDP can lead to an increase in the current prices of stocks, commodities, real estate, gold, art, etc. But this doesn’t occur because money is a medium of exchange. Suppose an auction house was silly enough to demand payment for multi-million dollar paintings in shares of Apple stock. That might add 1% to the normal transactions cost for a wealthy investor buying a Van Gogh, but wouldn’t dramatically affect the price of the $100 million painting. If the price of Van Gogh’s was soaring because an easy money policy was quickly raising future expected NGDP, those paintings would soar in value even if not paid for with money. (Whether money raises NGDP via its medium of exchange role is a tougher question, where even market monetarists are split.)
8. If wages and prices were 100% flexible then money would have almost no real effects. Because wages and prices are sticky, money does have real effects on output and real asset prices. But not because money “pours into” particular asset markets, rather because more money raises NGDP, and this raises real output (if wages and prices are sticky), which raises real asset prices.
Money matters, but not for the reason most people believe.
Saving matters, but not for the reason most people believe.
Money and saving affect the business cycle (if at all), by affecting either M or V.
Tags:
20. January 2012 at 09:26
Another simple reason why money cannot pour into an asset class is that for every buyer there is a seller. The cash spent on purchasing an asset is the same cash received by sellers of that asset.
20. January 2012 at 09:43
I had always though that money “pouring” into an asset class, say X, meant that lots of people have decided to trade other assets for X. So they are now liquidating other assets and redeeming that money for X. It connotes some change in behavior as well as large flows of money into the market. Isn’t that the traditional interpretation?
20. January 2012 at 09:45
National saving is personal saving plus business saving plus government saving. I don’t know exactly what term to use for aggregate saving as oppose to individual saving, but I think national saving isn’t the right one to use. Also, we aren’t really in a closed economy, and so national saving doesn’t equal national investment. Your argument about saving as a whole and investment is right, but applies to the whole world.
I know you realize the part about the whole world (and should have clarified the closed economy assumption.) My point is just about terminology.
And I suppose that in a closed economy national saving does equal investment.
But..
personal saving plus business saving = private saving
private saving plus government saving = national saving.
Investment equals national saving plus the net capital inflow.
Right on the money about money going into markets.
However, if the demand for money (in nominal terms) is a function of nominal wealth, then higher asset prices raise the demand for money. So, if the quantity of money rises, and the prices of stocks rise, which also raises the demand to hold money, then.. well, that is the sense in which the new money is in the stock market. People are willing to hold an increased stock of money because stock prices are high.
If we think about the disequilibirum, where there is initially an excess supply of money, and we suppose those with excess money balances buy stock, then that tends to raise stock prices.
If higher stock prices raise nominal wealth and so the demand to hold money (in nominal terms,) then we return to equilibirum. While the new money is not literally in the stock market, it is the higher stock prices that raised nominal wealth and raised nominal money demand, and cleared up the excess supply of money that started the process.
I agree that this process will tend to cause increases in the nominal demand for consumer goods and capital goods–nominal GDP. And that is what I think monetary insitutions should worry about, but the above is my argument regarding monetary disequilibirum and asset prices. It could be an important factor.
20. January 2012 at 10:16
Great post Scott.
I think it is useful, too, to imagine variations on your new house scenario in point 3.
What if, instead of buying a house, I decide to fulfill my lifelong dream of rafting the Grand Canyon? It will take me a decade to pay off my credit card, but I spend $20k to take my whole family on a specially-chartered rafting trip.
No capital goods are created, but the rafting outfitter/guide’s bank account just swelled by something close to $20k. By tapping my capacity to borrow, I simultaneously created $20k of additional debt and $20k of additional deposits in the system.
20. January 2012 at 10:35
One last go-round with the balanced budget multiplier, using your language this time:
Now do the same example with a new *bridge*. *The government taxes* $200,000 out and buys a new *bridge*. Same as before. The man who builds the *bridge* gets $200,000, so he actually increases his new saving by $200,000. Aggregate saving in the economy goes up by $200,000. What’s different is that the new *bridge* didn’t exist before being built, hence total wealth rises by $200,000. Following the money is completely unhelpful for thinking about national saving, all that matters is whether new capital goods have been produced.
I still think the Keynesian proof is simpler.
20. January 2012 at 10:36
“I have the same problem with people who talk about money being “saved.” At the individual level that’s true, but my decision to hold on to more dollars means someone else is holding on to less dollars. If we (collectively) increase our real holdings of base money, that is contractionary for NGDP.”
That is, there is a “paradox of thrift” and if the private sector is engaging in this kind of behavior, and we don’t want NGDP to fall, wouldn’t the solution be for someone in the economy to be less thrifty?
20. January 2012 at 10:46
DR:
The paradox of thrift is just a confusing way of describing the fundamental proposition of monetary theory. The individual can adjust his actual money holdings to desired money holdings. For the whole economy, given the quantity of money, the demand for money must adjust to the given quantity.
Saving is irrelevant. As long as people save some way other than accumulating more money balances, or else, if the do save by accumulating money balances, the quantity of money rises to match that demand for money, then there is no problem.
Finally, discouraging thrift is not really the answer. The goal should be to coordinate thrift (saving) and investment (the purchase, production, and sale of new capital goods.) The price that does this is the interest rate, and it does so partly by discouraging (or encouraging) thrift and partly by encouraging (or discouraging) investment.
Again, to the degree people choose to save by accumulating money, and the quantity of money is given, then the interest rate will fail to provide this coordinating role.
If, on the other hand, the quantity of money adjusts to meet the demand to hold money, then the composition of demand and the interest rate take care of themselves, and so, Sumner suggests we can ignore these things.
20. January 2012 at 10:53
Bank reserves stayed the same but leverage was allowed to go from 8 to 1 to somewhere around 30-40 to 1.
New “value” is created on the seller side, just like the builder of a new home, when the next transaction is at a higher price. Leading to a higher asset base. See it on the flow of funds statement. Right or wrong when that asset value went away it triggered deflation.
Because no one kept the nominal levels high enough to support the “value” in current properties.
20. January 2012 at 11:25
window washer:
You are confused. Leverage has to do with the ratio of bank assets to bank capital. Bank reserves are a types of assets–vault cash and reserve balances. The vault cash is used to provide for currency needs and the reserve balances are used for settling interbank clearings. Well, these days reserve balances serve as a bank investment. U.S banks are subject to reserve requirements based upon reported levels of transacton deposits. Transactions deposits are small fraction of total bank liabilities. Savings accounts are the biggest and CDs are second.
Investment banks, which used to hold lots of mortgage backed securities, had no reserve requirements at all, did not hold balances at the Fed, and didn’t have much “vault cash.” While they were very leveraged, with assets much larger than capital (and, in fact, much more leveraged than the commercial banks,) bank reserves had little to do with their activity.
20. January 2012 at 11:42
I think as Bill Woolsey, that money first induce asset-prices movements – if I understand his argument.
In any case The straight link between money and NGDP does not convince me, in the sense that money demand is determined in relation not only to real markets. Asset-prices are -I think- more important in the short term. There is a lot of financial transactions do not included in NGDP, that do not creaste capital but that use money.
For instance, the house bubble. As for te idea of Bill that assets-prices rises tend to rise money demand, I suppose it depends on expectations.
20. January 2012 at 12:17
“4. Others point to the money created by the Fed, often called the monetary base. Prior to mid-2008, this was composed of non-interest-bearing cash plus n0n-interest-bearing reserves. Some argue that an easy money policy in the 2000s tended to boost real estate prices. But why would this be true? One possibility is that the Fed put lots of reserves into the banking system, and these reserves were somehow “lent out” for new home purchases. In fact, roughly 95% of all new base money created in the years leading up to until mid-2008 was currency that went into people’s wallets, not new bank reserves. This is even true in the short run. Monthly increases in the base are overwhelmingly currency, with just a little bit of extra bank reserves (until late-2008.) So it’s hard to see how this new “money” could have been pouring into the housing markets. Money doesn’t have to “go somewhere,” 95% of it stays in wallets (prior to 2008).
It’s the credit expansion from the banking system that “poured into” the housing market. The banks expanded loans in the trillions during the housing boom, and that new money, which could not have been created had the Fed not increased bank reserves and not stood by with their “Greenspan put”, found its way into loans to mortgage buyers. So yes, new money “poured into” the housing market by way of new home buyers taking out easy money loans from the banking system that did not exist before.
“5. Others argue that the Fed held interest rates to very low levels. But how did they do this? Interest rates are set by the markets, not the Fed. They can influence interest rates by injecting money into the economy, but if money was very easy shouldn’t we have observed a very rapid growth in NGDP? Or at least a big increase in bank reserves, if you prefer that metric? How do we know that rates weren’t low simply because the supply of credit schedule shifted right faster than the demand for credit schedule? Why assume monetary policy is involved at all?”
If it wasn’t credit expansion that did it, but simply a change in demand schedules, then as the boom in house prices continued, and as mortgage backed securities demand increased, there should have been a fall in spending and hence fall in prices elsewhere.
Greenspan and Bernanke blamed the boom on a “savings glut”, from China and other countries. But then if it was a boom in savings that did it (shift in demand schedules), then consumer prices and consumer profits should have drastically fallen. But in reality they did not. They not only did not fall, but they rose as well. Profits in general were booming. Therefore, the ONLY explanation for this is an increase in the quantity of money and volume of spending. If you don’t see NGDP booming, then it’s not the above theory that is flawed, it’s the NGDP methodology that is flawed. It’s the same flawed reasoning that prevented economists from identifying a federal reserve induced boom during the 1920s, on the basis that consumer prices were more or less stable.
More importantly, it’s not so much the quantity of new credit and money supply as it is the resultant discoordination between investors and consumers brought about by artificially low interest rates that explains the Fed messing up the economy. Yes, NGDP could remain at a “stable” growth rate, al nice and clean on paper, but the economy of PEOPLE is becoming more and more distorted over time, as investors continue to make bad decisions that are not sustainable because the real capital isn’t there. At some point, what is invested, and relative, are going to be revealed as not in line with true consumer preferences.
This game of passing the buck, in order to salvage one’s ideal policy prescriptions, is a huge problem in economics.
“6. Prior to 2008, the amount of new money created by the Fed is typically trivial, compared to the size of the economy. And it was usually injected by buying T-securities, which is a huge and liquid market. Thus it’s hard for me to see how the so-called Cantillon effects were important. On a typical day the Chinese would buy more Treasury debt than the Fed. If the Fed had bought German and Japanese debt instead of T-securities, then the holders of those securities would switch over and buy the T-securities the Fed was no longer buying. The net effects would be tiny.”
It’s interesting how you feel compelled to minimize the importance of Cantillon effects, like it’s some stingy family member who takes away the NGDP party punch bowl.
http://research.stlouisfed.org/fred2/series/M3
The Cantillon effects from the mid-1990s to 2006 were HUGE. $4 trillion to over $10 trillion (in M3) means that there was a total of some $6 trillion created out of thin air in a little over 10 years on top of $4 trillion, that by the nature of the monetary system went to some individuals before other individuals.
Sure, sometimes it was the primary dealers through US debt monetization who received new money first (actually it was almost ALWAYS the primary dealers first), sometimes it was home buyers through credit expansion from commercial banks who received new money first. But in total, there was $6 trillion that generated Cantillon effects. Compare that to the original $4 trillion M3 in the mid-1990s, and I defy anyone to truly label this $6 trillion from the Federal Reserve System as creating Cantillon effects that were “minimal.”
“7. Monetary policy can have a big effect on asset prices, but the effect has little to do with what assets the Fed is buying (until 2008, after that the purchases were so massive it might have marginally shifted interest rate spreads.) My hunch is that if the Fed had bought index stock funds during 2002-06 instead of T-securities, the impact on asset prices would have been roughly the same. If there was an impact, it was because monetary injections can raise expected future NGDP, and higher expected future NGDP can lead to an increase in the current prices of stocks, commodities, real estate, gold, art, etc. But this doesn’t occur because money is a medium of exchange. Suppose an auction house was silly enough to demand payment for multi-million dollar paintings in shares of Apple stock. That might add 1% to the normal transactions cost for a wealthy investor buying a Van Gogh, but wouldn’t dramatically affect the price of the $100 million painting. If the price of Van Gogh’s was soaring because an easy money policy was quickly raising future expected NGDP, those paintings would soar in value even if not paid for with money. (Whether money raises NGDP via its medium of exchange role is a tougher question, where even market monetarists are split.)”
It’s not the short term t-bills, or the effect on short term interest rates, that is the decisive factor when talking about monetary policy and the housing boom. It’s how the Federal Reserve brings down short term rates that is decisive. They do so by creating new bank reserves. These reserves are what lay the foundation for fractional reserve banks expanding credit with maturities of all types, including long term rates in mortgage lending.
In your example of the auction house, if it is to represent the whole economy, then one can’t claim that higher future inflation expectations can necessarily boost all prices today concomitant with it. But current prices are a function of CURRENT money supply, not future money supply. Even if one were to argue that M3 will rise 10,000% tomorrow, and everyone believed that person, then we still couldn’t bring about an increase in prices we pay today concomitant with that expectation, because everyone is cash constrained the current money supply. We can only bid prices up to the amount of money we have today, not how much we think we might have tomorrow. Increasing inflation expectations can only have a delimited effect on prices today, through a lower demand for money for holding. But people can’t hold more money in the aggregate, and they certainly can’t spend more of their money faster if there isn’t an equivalent rise in the speed of production and delivery of goods and services. If I can only produce 1 widget per day, then you won’t be able to spend your money any faster than once per day on me.
20. January 2012 at 12:35
OK, I did not know that that had a name: Cantillon effect. So asset bubbles are Cantillon effects. Good.
20. January 2012 at 12:52
My remarks may simply reflect irremediable confusion about economics, in which case please do not feel obliged to answer them. Still . . . .
You write, “my decision to hold on to more dollars means someone else is holding on to less dollars.” It seems that you are (strangely) assuming a constant money supply. If the quantity of money–the number of dollars–is increasing, I can hold onto more while no one else is holding onto less.
“If we (collectively) increase our real holdings of base money, that is contractionary for NGDP.” Further explanation would be appreciated. To say that we collectively increase our real holding of base money is just to say that the real quantity of base money (in private hands?–I’m not sure who “we” are) increases. I suppose this has happened over time; surely the real quantity of base money in the U.S. is greater now than it was 100 years ago. But why think this *contractionary*? Would NGDP be higher if it hadn’t happened?
(These questions are aside from your main points, which seem obviously sound.)
20. January 2012 at 13:23
Scott,
Very illuminating as always. Perhaps you have addressed this before as I am new to your blog, but what in your mind caused the sudden collapse of NGDP in mid 2008? Was it caused by specific actions taken by the Fed? Was there a collapse in the availability of credit? Was there a sudden contraction in the money supply because the “money” had vanished due to the collapse of credit “acting like money” i.e. syndicated loans, wholesale markets, etc? Or did some intervening event lead to a rapid deceleration in the velocity of money? Maybe the collapse in oil prices led to a sudden onset (but why?) of Fisherian debt deflation?
20. January 2012 at 13:38
OneEyedMan, That’s right.
Neal, I think that’s the traditional view, but it’s wrong. When assets are bought by one person, they are sold by another.
Bill, I agree about saving. I define money as the base, and I doubt the demand for base money goes up when stock prices rise. Indeed in my case I’d probably hold less base money, because a stock boom is usually associated with higher interest rates.
Brett, Good example.
Benjamin, You are confusing accounting with causality (I think, I’m actually not sure your exact point.)
If that’s all that happens then wealth might go up as you say. But what if the taxes required to pay for the bridge crowd out an equal amount of private investment.
DR, No, in that case the problem isn’t thrift, it’s money hoarding. Have people save all they want, as long as it isn’t hoarding cash.
The Window Washer, I certainly think that’s a big part of what happened.
Luis, See my response to Bill.
Major Freedom, You said;
“It’s the credit expansion from the banking system that “poured into” the housing market. The banks expanded loans in the trillions during the housing boom, and that new money, which could not have been created had the Fed not increased bank reserves and not stood by with their “Greenspan put”, found its way into loans to mortgage buyers. So yes, new money “poured into” the housing market by way of new home buyers taking out easy money loans from the banking system that did not exist before.”
You are confusing money and credit. Banks can expand credit without any additional money from the Fed.
Regarding point 5, why would prices fall, if the Fed was targeting CPI inflation at about 2%?
6. M3 is mostly composed of credit, I’m talking about money created by the Fed, not credit. There wasn’t much money creation by the Fed.
7. You said;
“They do so by creating new bank reserves.”
But they only created a tiny amount of bank reserves. Mostly they were creating currency held by the public.
You said;
“But current prices are a function of CURRENT money supply, not future money supply.”
Actually current prices are much more influenced by the expected future money supply. Suppose you were about to buy a house for $500,000. And you suddenly heard that the Fed planned to reduce the money supply by 90% next year. Might you wait, and try to pick up the house for $50,000 next year? If so, wouldn’t the price fall immediately?
Luis, No, Cantillon effect refers to the impact on monetary policy on a particular asset class. You can have bubbles without any change in monetary policy.
20. January 2012 at 13:48
Philo, I assume that when I decide to hold more dollars it doesn’t cause the total supply to increase. (which is admittedly debatable, but useful for analytical purposes.)
You asked;
“But why think this *contractionary*? Would NGDP be higher if it hadn’t happened?”
Yes, much higher. The money supply has soared over the past 200 years. If the real demand for money hadn’t changed, then the price level would have soared just as fast as the money supply. And NGDP is highly correlated with the price level in the long run (given 3% trend RGDP growth.)
Michael, Very complicated question that I can’t answer quickly. The monetary base grew only about 1% in the 12 months to May 2008. That slowed NGDP growth. Then the bursting of the housing bubble and high oil prices sharply reduced the demand for cars, trucks and houses. This sharply lowered credit demand in April to October 2008. So the Wicksellian equilibrium rate fell sharply, but the Fed kept it’s target rate at 2% during that period. This made monetary policy effectively much tighter, and then when the financial crisis got bad the demand for liquidity increased sharply, and the Fed made the mistake of paying interest on reserves, which further increased the demand for liquidity. Lots more could be said, but those are some factors.
20. January 2012 at 14:05
The stock market goes up because the average person decides to hold a greater percentage of his wealth in stocks vs. other assets. This can happen without any transactions at all; one person sits around with a smug look on his face, while the other says $#!+ I wish I had bought yesterday.
20. January 2012 at 14:14
ssumner:
“It’s the credit expansion from the banking system that “poured into” the housing market. The banks expanded loans in the trillions during the housing boom, and that new money, which could not have been created had the Fed not increased bank reserves and not stood by with their “Greenspan put”, found its way into loans to mortgage buyers. So yes, new money “poured into” the housing market by way of new home buyers taking out easy money loans from the banking system that did not exist before.”
“You are confusing money and credit. Banks can expand credit without any additional money from the Fed.”
Credit is necessarily money. Money is not necessarily credit.
Yes, bank can expand credit money without additional base money from the Fed, but they can’t KEEP expanding credit money without additional base money from the Fed, and, by the same token, they can expand more credit money if the Fed gives them more base money.
“Regarding point 5, why would prices fall, if the Fed was targeting CPI inflation at about 2%?”
That’s exactly why they can blow up asset bubbles. They target 2% in CPI, and if the conditions are right, keeping a 2% CPI in the presence of a savings boom and housing boom that would have otherwise generated -2% CPI, will require HUGE quantities of additional credit money pyramided on top of additional supporting base money. This is because as the Federal Reserve System inflates, and people don’t allocate their funds in such a way as to generate a 2% CPI, but they speculate with it, like on housing, then the Fed will just keep easing money further and further, until the banks issue so much new credit that the Fed gets their 2% CPI. But by the time they do get it, there is already way too much money circulating in the capital goods and financial sectors, and so once the Fed eased, as they did starting in 2005, to avoid too high of a CPI, then all that capital and housing that depended on cheap money and was supported by cheap money, collapsed.
It’s not about aggregate spending, even if you measure a different statistic like NGDP. It’s the discoordination that is most important, not price levels.
“6. M3 is mostly composed of credit, I’m talking about money created by the Fed, not credit. There wasn’t much money creation by the Fed.”
Sure, but I am talking about credit money, because it DEPENDS on the base money creation from the Fed. Without the Fed backstopping commercial banks, through base money and “Greenspan put”, the banks could not have created that additional credit.
I said: “They do so by creating new bank reserves.”
You said: “But they only created a tiny amount of bank reserves. Mostly they were creating currency held by the public.”
Tiny? Base money almost DOUBLED from 1995-2005, from $400 billion to $800 billion!
http://research.stlouisfed.org/fred2/series/BASE
A doubled base money of $400 billion, coupled with the money multiplier, is capable of generating TRILLIONS of new dollars via credit.
How can you say this was tiny?
“But current prices are a function of CURRENT money supply, not future money supply.”
“Actually current prices are much more influenced by the expected future money supply. Suppose you were about to buy a house for $500,000. And you suddenly heard that the Fed planned to reduce the money supply by 90% next year. Might you wait, and try to pick up the house for $50,000 next year? If so, wouldn’t the price fall immediately?”
Sure, but then because I was standing ready with $500,000, it means that I now have $450,000 to spend elsewhere, which means prices for other things will rise.
For GENERAL increases and decreases in prices, that is primarily a function of current money supply, not future money supply.
Suppose in your example you heard that the Fed is suddenly going to increase the money supply by 90% next year. How can you increase your demand for the house you were going to buy if you only had $500,000 to begin with? The price of that house can’t rise, because in the present, you only have $500,000.
20. January 2012 at 16:32
Here’s a sentence I never expected to come out of Scott Sumner’s keyboard:
Money and saving affect the business cycle (if at all), by affecting either M or V.
Scott, what’s with the “(if at all)”? Now you’re not so sure that money affects the business cycle? I thought that was your reason for living?
And are you saying there are people out there, who think money affects the business cycle but *not* through M? What would that even mean?
20. January 2012 at 17:28
“Monthly increases in the base are overwhelmingly currency, with just a little bit of extra bank reserves (until late-2008.) So it’s hard to see how this new “money” could have been pouring into the housing markets. Money doesn’t have to “go somewhere,” 95% of it stays in wallets (prior to 2008).”
I have always wondered about the $900 billion or so in US cash in circulation, or about $3,000 for every resident of the USA, including children and my cheap relatives. No one carries cash like that, obviously.
Most currency is offshore somewhere, and usually thought to facilitate black markets (usually drug lords).
What does this mean for monetary policy? Trade policy? I find economists detest these questions. I hope someday Sumner addresses these questions.
Lastly, we hear a lot of honking about “money is a store of value,” etc. For who? Drug lords? Our monetary policy to make sure drug lords do not suffer inflation?
20. January 2012 at 17:32
Interested in Bob’s question too.
20. January 2012 at 18:01
Bill Woolsey
“Investment banks, which used to hold lots of mortgage backed securities, had no reserve requirements at all, did not hold balances at the Fed, and didn’t have much “vault cash.””
Citi, JP Morgan, Wells Fargo, Bank of America ec.. dont have reserve requirements?
Scot’s blog is a great place to be wonky as this post shows. I just like to shoot a “Um Hello.I the real world….” reminder into the mix.
Really what I was pointing out is that “balance sheet” games are very important to the issues Scott writes about.
Scott got that. I’m not much of a writer but the amazing think about him it that almost everytime I’ve shot something out over the last couple years he got what I meant.
20. January 2012 at 18:24
bob
“Scott, what’s with the “(if at all)”? Now you’re not so sure that money affects the business cycle? I thought that was your reason for living?”
My thought is
Scott doesn’t thing you can tame the business cycle, unlike Greenspan. He thinks it’s better to buffer it’s effects in downturns instead of strangling it on the way up(Greenspan). NGDP targeting is what he thinks will do that.
Benjamin
“Most currency is offshore somewhere,”
The reason that 2/3 of our currency is outside the country is it’s more trusted and stable than local currencys. Our money supply is better than the local.
and usually thought to facilitate black markets (usually drug lords).
A few percent of the overseas currency is drug lord ec.Mostly it’s a store of value or business transactions.
A larger portion of the currency is used in the US black market. Ever asked for a different price for paying cash? How much of the drug market mark up is in the US and never crosses the border? 80-90%
“What does this mean for monetary policy? Trade policy? I find economists detest these questions. I hope someday Sumner addresses these questions. ”
Currency supply has nothing to do with monetary issues. If it did all those dollar coins would be in someones pocket.
20. January 2012 at 18:42
Window Washer–I think Scott will have a different answer.
On US cash overseas: My understanding US cash is mostly in black markets–not used as a store of value, except by people who do not care about some slippage in value (drug lords).
I have traveled quite a bit and while people will take dollars, that is usually because they are offering you an exchange rate below the bank or official rate. They plan to take your dollars and convert to local currency.
Anyway, about three-quarters of US cash in circulation is in $100 bills. Really, you are plying cantinas in El Salvador with $100 bills? They are storing $100 bills? Why is there such a large amount of $100 bills in circulation?
I think it is obvious that $100 bills are used for black market transactions, where the dollar volumes are high. Drug deals.
This is where the gold-nuts and paper-currency fetishists get crazy. If you say we should expand the money supply, they will say that is theft from existing currency holders. Dilution.
But if drug lords were to repatriate their US dollars, that’s okay. Of course, such a huge increase in circulation in the USA would boom the economy, and lead to moderate inflation–there would be dilution also. But morally, it would be okay.
Because it is immoral to juice the money supply to bring about prosperity. It is better to languish for generations, like Japan.
Here’s another downer for you:
http://www.clevelandfed.org/research/data/inflation_expectations/index.cfm
Inflationary expectations have never been so low. Why are fighting inflation now?
20. January 2012 at 20:09
Dear Dr. Sumner,
While I agree with the principles behind your post, I find most of times the metaphors send better signals for interpretation. But my main point is about prices and wages.
Prices and wages are embodiment of information – historical, present and future. If they change too quickly then the historical aspect is lost. For better or worse human scale of value are anchored to the past. It gets influenced by present and to a less degree by expectations of future. If we lose our anchor point or the reference scale then our mental models collapse and we lose our sense of reality.
Then comes the question of pouring. What pouring refers to is change in the relative value of asset classes. Imagine a spread of assets along a value spectrum, sort of a hierarchy (with sometimes assets jointly occupying a hierarchal position).
If money increase does not modify the hierarchy then it does not impact much. If it does then it creates gainers and losers just because money is created. For example, if a really thirsty person would rather be just under the tap than away where water will eventually get to him
The argument therefore is whether the government or central bankers be allowed to create such distortion that has no grounding of productivity or real value creation.
One can argue that over long term the asset hierarchy goes back to a certain mean. But during the time a distortion is set in motion and the time we get back to time-tested mean we can extract advantage. Finance is prepping to do just that.
Rahul
20. January 2012 at 20:41
Thanks for your answers, one of which left me still puzzled. You wrote: “I assume that when I decide to hold more dollars it doesn’t cause the total supply to increase.” My comment said nothing about causation (nor did your original post). My point was that so long as the quantity of money is increasing (regardless of cause), one person can hold more while no one else holds less. Thus your original statement needs a rider, “. . . so long as the total quantity of money is unchanged” (or the like.)
You also wrote: “You asked; ‘But why think this *contractionary*? Would NGDP be higher if it hadn’t happened?’ Yes, much higher. The money supply has soared over the past 200 years. If the real demand for money hadn’t changed, then the price level would have soared just as fast as the money supply. And NGDP is highly correlated with the price level in the long run (given 3% trend RGDP growth.)” Now I understand. I was thinking of ‘contractionary’ in real terms; you meant it in nominal terms. (How is it usually used in common parlance? Isn’t one of your themes that people often confuse real and nominal magnitudes? Doesn’t common economic terminology foster such confusion?)
20. January 2012 at 21:09
Consider two different houses, built via $200,000 in spending:
(1) House #1 is built in a week, and is designed to last no more than half a year.
(2) House #2 is built in a year, and is designed to last at least 200 years.
It’s ridiculous to say that total wealth has gone up the same in each case (and we don’t need to know why such choices were made — let’s not change the subject please).
As you say, following $$$ spent is no way to think about savings ..
And NOTE WELL: Following the money give us nothing but a past measure — a pre-marginalist, add-it-up, backward looking measure of value.
The valuation of production goods is always forward looking (sunk costs and all the other stuff Wieser and Jevons taught economists).
And the same goods bought with the same number of dollars can be used in any number of alternative production processes, some of them longer and producing superior output, and some of them shorter and producing inferior output (we don’t have to consider the other alternatives, because those do not pose the same problem of choice / valuation — we will only produce longer production processes if these processes promise superior output, otherwise we won’t even consider them).
That is, exactly the SAME “S” savings measured in dollars can provide us with alternative production paths capable of producing radically different output: if we are willing to sacrifice time and a bit of consumption in the intermediate term we can have superior output, if we are not willing to sacrifice time and a bit of intermediate consumption, we will be left with inferior output.
In other words the value of “I” is contingent our our alternative production choices NOW across time in the future. _NOT_ dollar volume, not physical volume, but choices between alternative production _uses_.
Scott writes,
“The man who builds the house gets $200,000, so he actually increases his new saving by $200,000. Aggregate saving in the economy goes up by $200,000. What’s different is that the new house didn’t exist before being built, hence total wealth rises by $200,000. Following the money is completely unhelpful for thinking about national saving, all that matters is whether new capital goods have been produced.”
20. January 2012 at 22:59
Benjamin
“I have traveled quite a bit and while people will take dollars, that is usually because they are offering you an exchange rate below the bank or official rate.”
I usually get a better cost basis on the “street”.
“They plan to take your dollars and convert to local currency.
”
Quite a presumption.
“Anyway, about three-quarters of US cash in circulation is in $100 bills. Really, you are plying cantinas in El Salvador with $100 bills?”
El Salvador it was a middle of no where road side cantina that changed my $100. 30 miles past the border I still didn’t have local currency, paid the boarder charges in dollars also. No bribes involved. Not making this up.
It’s easier to change $100 bill in a 3rd world corner store than one in the US.
“They are storing $100 bills?”
Yes.
20 years ago I stood in the line of an Uzbek bank around 15 people holding $5-15 thousand in there hands in front of everyone including the police at the entry. I doubt that was black market money.I’ve lost track of the number of contries I’ve been in, maybe changed money 10 times in a bank, 100 times on the street and 500 times in a business.
“Why is there such a large amount of $100 bills in circulation?”
Demand.
“I think it is obvious that $100 bills are used for black market transactions, where the dollar volumes are high. Drug deals.”
Yes $100 are used in the black market.
Most “black market transactions” outside the US are less than $100. Most drug deals are less than $100.
There is a lot of infomation on currency use your theories are simply wrong.
Don’t like drug lord? Fine, it has nothing to do with currency supply.
In May 2008 I pulled $52,000 out of the bank and into currency. The teller ask why I didn’t want to get some kind of “investment” return on it instead of holding it in cash. I looked at him and said “Cash is an investment.” I took the tax hit and closed out my IRA in July 2008.
I made cash purchases of assets over the next year that all yeild over 30%.
It’s all on my 1040’s
I can close land purchase within 12 hours on a weekend with a 3am closing at a title company because of currency.
I know people holding millions in currency.
20. January 2012 at 23:07
Greg
Great example.
Most tract home builders use a 20-30 year architectural standard.
Most of my house has 100+ standard. Damn roof will be shot in 50-70 years.
Monday I’ll working on a house being built to a 600 year architectural standard.
Yet they all show up the same in an equation and GDP.
Now guess the leverage amount on each and think of the momentum of money.
20. January 2012 at 23:14
Humility: n. rereading posts that don’t allow spell/gramar check.
Homage: v. lifting your beer to Bill for the F7 key.
Or as I say “Let see what Bill says.” before punching the F7
21. January 2012 at 05:03
[…] Sumner against Austrian macro. […]
21. January 2012 at 05:32
window washer:
Citibank, Bank of America, and Wells Fargo and commercial banks with reserve requirements on transaction balances.
Before 2008, Goldman Sachs, Lehman Brothers, and Morgan Stanley were stand-alone investment banks. They had no transactions accounts or reserve requirments.
All of them now have commerical bank charters, can issue transactions accounts,and have reserve requirements for those.
Capital = net worth. Leverage or capital ratios looks at the ratio of net worth to total assets. How much total assets are funded by the stockholders vs. by debt, the banks creditors.
Reserves = vault cash plus reserve balances at the Fed. Reserve requrements relate how much reserves a bank has relative to how much transaction deposits its customers have.
There isn’t much connection between the two.
In finance, the reserve requirement is pretty much ignored because it is trivial. They focus on leverage.
For anti-banking types on the internet–partcularly amateur austrians, reserve requirements are emphasized. They often confuse leverage and the money multipler.
If all banks are commercial banks and they only offer checking accounts, reserve requirements would be much more important to bank lending. Still, capital and leverage would still be different. It would relate owners equity to total bank assets–more or less bank lending.
21. January 2012 at 07:14
woolsey, you mis-spoke,
Under current conditions, wanting less leverage doesn’t make you an anti-banking type.
It makes you sane.
Toss FDIC and TBTF and banks can do whatever the hell they want.
The end result is the same – the cost of borrowing goes up, and so does capital accumulation.
spend some time looking at DWOLLA.
Then imagine if we are forced to keep FDIC, we make it very, very easy for Internet start ups to start a FDIC backed virtual bank.
There are MANY MANY way to skin this cat woolsey, I’m not anti-banking or anti-education – I’m an Internet guy and I want to eat their cheese.
21. January 2012 at 09:15
————-quote————-
(1) House #1 is built in a week, and is designed to last no more than half a year.
(2) House #2 is built in a year, and is designed to last at least 200 years.
————endquote———–
Hard to believe they’d both sell for the same amount.
21. January 2012 at 10:01
I am not sure if this is a quibble or something is going on that I do not understand, but in the last example in Prof. Sumner’s point (3), I would think that total wealth created by selling a new $200K house would be decreased by the builder’s expenses. E.g. if the builder had to spend $150K to build the $200K house he sells, total wealth increases by $50K, not $200K.
21. January 2012 at 11:38
Major Freedom. Good chart on BASE from 1985 to 2008, but even more amazing is why no hyperinflation after 2008 when BASE went to the moon?
http://research.stlouisfed.org/fred2/series/BASE
21. January 2012 at 11:41
Scott,
It seems to me the important mechanism underlying changes in MV (i.e. NGDP) is credit. M is mostly made up of the broader monetary aggregates and changes to M are mostly the result of changes to the amount of credit in the economy. So when you think about how events or actions impact NDGP, you need to focus on how they impact credit. Expectations and asset prices have the biggest impact on the amount of credit in the system.
Another way of thinking about this is leverage. Increasing the leverage (asset to equity ratio) of financial institutions increases M and thus NDGP. When asset prices drop you get a de-leveraging and NDGP falls.
21. January 2012 at 11:45
Steve, That’s right.
Major Freedom, No, credit is not money. If my local banker loans me $100, he can’t take that loan to Walmart and buy something.
You said;
“You said: “But they only created a tiny amount of bank reserves. Mostly they were creating currency held by the public.”
Tiny? Base money almost DOUBLED from 1995-2005, from $400 billion to $800 billion!”
I was clearly talking about reserves, not the base.
You said;
“How can you increase your demand for the house you were going to buy if you only had $500,000 to begin with? The price of that house can’t rise, because in the present, you only have $500,000.
Suppose Bill Gates has no money at all. But he has $30 billion in wealth. Do you really think he’d be unable to buy a home? You don’t need money to buy homes, you need wealth. If the home is expected to be worth much more in the future, that will make it worth more right now.
Bob, Not every change in money affects the business cycle. And monetary policy can also affect V.
Ben, Good point. Most money is held to avoid taxes. A slightly faster rate of inflation and NGDP growth would hurt cash holders the most (others are partially protected via the Fisher effect.) And those victims are mostly people breaking the law.
The Window Washer, The large amount of cash hoarded in the underground economy (much of which is in the US, much of which is not) has surprisingly little effect on monetary policy, as the Fed passively accommodates that demand so that interest rates and inflation are pretty much unaffected. If they didn’t (say under a gold standard) the effect of cash hoarding would be deflationary.
Ben You are right about the $100s, it’s mostly a form of saving, not transactions balances.
rahul, It’s true that monetary policy can distort asset values, as when gold soared in the late 1970s. But the housing boom was not caused by monetary policy, contrary to what people claim. That was one reason I did this post.
Philo, Monetary policy matters because it creates nominal shocks. Since lots of contracts are in nominal terms, that distorts the real economy.
Greg; You said;
“Consider two different houses, built via $200,000 in spending:
(1) House #1 is built in a week, and is designed to last no more than half a year.
(2) House #2 is built in a year, and is designed to last at least 200 years.
It’s ridiculous to say that total wealth has gone up the same in each case (and we don’t need to know why such choices were made “” let’s not change the subject please).”
It may be ridiculous, but I believe it to be true. Let me ask you this. Suppose the less durable house was $210,000 and the sturdy house was $200,000. And someone offered to give you either house. Which would you take? I’d take the house that cost $210,000 and was going to quickly fall apart, and then I’d sell it for cash.
Values are determined in the marketplace.
pct, Lets say someone else produced the wood for the house and sold it to the builder for $150,000. Then that person’s wealth goes up by $150,000 and the builder’s wealth goes up by $50,000
21. January 2012 at 11:48
dtoh, I define ‘money’ as the monetary base, so credit plays no role in my analysis. Only the Fed can create base money, not the banks.
21. January 2012 at 12:47
Scott, you say
“I define ‘money’ as the monetary base, so credit plays no role in my analysis. Only the Fed can create base money, not the banks.”
I understand that well, but clearly M (of MV) is more than just base money, and the banks can and do create M with or without any Fed impetus. If this is correct then what you are saying is that the mechanism by which base money impacts MV is not important to the analysis. Am I reading you correctly?
21. January 2012 at 18:33
As a follow-up to my last comment: If I had read your statement more carefully, I would have noted that you said “contractionary for *N*GDP.” My apologies.
But I am still grappling with your statement, which implies that less growth in the real money supply leads to more growth in NGDP. I wondered what the U.S. economy would have looked like if (say) the real quantity of money had remained unchanged over a long period, such as the last hundred years. You (rightly) claimed that, if the nominal quantity had increased *as it actually did* and *various other factors had been as they actually were*, the price level and NGDP would now have been much higher than they actually are. But what, exactly, are those *various other factors* that are being held constant, so as to generate this result? And why would the nominal quantity of money have grown as it actually did if the real quantity had remained constant? If we suppose that the nominal quantity might also have remained constant (because of some linkage between it and the real quantity), we must grant that NGDP might have been *lower* than in actuality.
I am still struggling to fill in the unstated background so as to understand your claim that a lower real quantity of money would be (nominally) contractionary.
21. January 2012 at 19:36
I thought I was rather polite in asking that the subject not be changed ..
Basically, you are agreeing with me, but don’t seem to know it.
Scott writes,
“It may be ridiculous, but I believe it to be true. Let me ask you this. Suppose the less durable house was $210,000 and the sturdy house was $200,000. And someone offered to give you either house. Which would you take? I’d take the house that cost $210,000 and was going to quickly fall apart, and then I’d sell it for cash.”
21. January 2012 at 19:45
You are assuming a constructed world where demand is “given” — it’s a fake world and isn’t the real world.
There is no “given” price to things, no guarantees.
HP bought Palm its WebOS for billions and is now giving WebOS away for free, and is giving away it HP Tablet for less than half the cost of making one.
In my neighborhood as house sold for $1,000,000 and then months later couldn’t be sold for $700,000. “Supposing” the price/value of something as a “given” in an economists fake example isn’t a reflection of the real world.
It’s this game which macroeconomists exploit to return to a retrograde Ricardian non-marginalist theory of value, where past measure in $$$ terms are pretended to be the same as marginalist relative valuations looking toward the future.
Scott writes,
“It may be ridiculous, but I believe it to be true. Let me ask you this. Suppose the less durable house was $210,000 and the sturdy house was $200,000. And someone offered to give you either house. Which would you take? I’d take the house that cost $210,000 and was going to quickly fall apart, and then I’d sell it for cash.”
21. January 2012 at 20:24
Yes, money doesn’t necessarily “pour” into markets out of whole cloth. Markets are vessels for moving money for either investment or consumption (most of the money going through oil markets eventually goes to fuel consumption). The movement of money through various markets may have very little to do with the monetary base.
However, that does not mean movements of money into and out of certain markets are a completely benign activity. If all markets efficiently allocate capital, then investment activity simply moves money to its productive use.
But what if capital is not efficiently allocated, through either principal-agent problems or outright moral hazard? For example, any way you look at it we built far more houses in Las Vegas and other markets than number of physical households. These houses were eventually bought by loans mostly held on bank balance sheets and which banks then funded mostly through the money market funds and short-term repo agreements. In turn, investors did not really evaluate default risk of these bank deposits because they had never comprehended these instruments defaulting.
If NGDP remains on track in such an environment, the mix of consumption and investment will tilt more to investment. Savers will think far too many instruments return acceptable interest and many of the investments from a credit bubble will not return enough money through future consumption.
After such a bubble pops, there is nothing stopping a central bank from still hitting their NGDP targets. I want to make that clear. The stock market only went down to the 6000’s because of a tight money spiral, not because capital was grossly allocated when stocks were at 13,000.
However, we are still stuck with five years of material and labor which were inefficiently used to build things the economy did not want. Big savers decided to lend money to the materials and labor to building a house in Las Vegas, instead of consuming the same material and labor because there were not enough good investment opportunities. Then when the investments default, the savers should ideally suffer the consequences of overinvesting.
The unfortunate fact, though, is that the taxpayers and regular retail savers are really left holding the bag, and not necessarily high net worth individuals. Without any bailouts after Lehman brothers, retial savers would have held the bag for the run on money market funds through both their personal holdings of money market funds and the holdings of money market funds in current accounts of corporate stocks and bonds they own. Repo agreements were also very popular with pensions and insurance companies, who have many retail stock and bond holders and whose liabilities are government-guaranteed.
In short, just because I/C balance should not affect NGDP targeting of a central bank does not mean a credit bubble or a big current account deficit has no ill effects. If the markets inefficiently allocate capital far more than a more regulated financial system would, then we should have a more regulated financial system to at least try to solve the inherent principal-agent problems in finance. And even if the government keeps NGDP on target, the credit misallocation will still hurt the standard of living of the savers/taxpayers who eventually hold the bag for such misallocations.
21. January 2012 at 21:18
Here’s another example where the $$$ expended measure assumed in “S= I” is pathological:
If any given firm which before used annually to buy a certain amount of machines in replacement of old ones decides to produce these machines in its own factories, the amount of money used to buy investment goods will decrease but the total amount of investment will not change.
21. January 2012 at 23:01
[…] Source […]
22. January 2012 at 09:14
dtoh, You said;
“Am I reading you correctly?”
No, when I use MV=PY, I mean the base, and base velocity. base velocity is NGDP/base.
Philo, You asked:
“I am still struggling to fill in the unstated background so as to understand your claim that a lower real quantity of money would be (nominally) contractionary.”
I don’t think I said this, can you provide a quotation?
Think of the money supply in the 1800s as being exogenous (gold discoveries, etc.)
Greg, I do not assume that the value of capital assets never changes, just that it’s possible for two houses with very different durability to have exactly the same market value at a point in time.
Matt, You said;
“However, that does not mean movements of money into and out of certain markets are a completely benign activity.”
I don’t agree that money goes into or out of markets, unless you believe money was pouring into the stock market on Oct. 19, 1987, when it crashed 22%. A record number of shares were bought that day, and the shares were bought with money.
I agree with most of the rest of your comment, but wouldn’t blame “the market” but rather really, really bad banking regulation, which encourages banks to engage in reckless risk taking.
Greg, You said;
“If any given firm which before used annually to buy a certain amount of machines in replacement of old ones decides to produce these machines in its own factories, the amount of money used to buy investment goods will decrease but the total amount of investment will not change.”
No, investment is not inter-firm purchases of capital goods, it’s funds spent on construction of capital goods.
22. January 2012 at 11:20
ssumner:
“Major Freedom, No, credit is not money. If my local banker loans me $100, he can’t take that loan to Walmart and buy something.”
100% absolutely false. Credit IS 100% money. That is why you can buy things with loans. That is why you can buy houses using a loan, why you can buy cars using a loan, why you can buy goods at Wal-Mart using your credit card. Just because you aren’t transferring physical currency, that doesn’t mean you are not giving them money. Once you buy a house using a loan, which is money, the home builder is finished dealing with you. He got his money. Now you owe the bank the principle of what the bank gave to the home builder, plus interest.
You said it yourself that credit makes up around 95% of M3. Well, M3 tracks money supply!
How can you say credit is not money? Credit is universally accepted as legal tender! If you bank loans you $100, then you can start to write checks and make payments to merchants off that loan. If loans are not money, then how the heck can people be buying things using loans, like houses, cars, and food, clothes, utility bills, etc on credit cards (which is also a loan)?
“Tiny? Base money almost DOUBLED from 1995-2005, from $400 billion to $800 billion!””
“I was clearly talking about reserves, not the base.”
OK, but then you are just guilty of ignoring the base, which is what the Fed does have control over, and is what underlies my argument that it is wrong to say the Fed had no part in the housing bubble.
“How can you increase your demand for the house you were going to buy if you only had $500,000 to begin with? The price of that house can’t rise, because in the present, you only have $500,000.”
“Suppose Bill Gates has no money at all. But he has $30 billion in wealth. Do you really think he’d be unable to buy a home?”
If Bill Gates has no money, but $30 billion in wealth, then that means he is able to get no more than $30 billion in cash for his wealth. He can’t get $300 billion, because the money supply is constrained to what it currently is such that his physical wealth is worth $30 billion.
Merely increasing the amount of wealth or money someone has from $500,000 to $30 billion, doesn’t change the principle of the argument that today’s prices are constrained to today’s money supply, not tomorrow’s money supply.
So instead of someone with a $500,000 loan considering buying a house for $500,000,or has zero cash but wealth totalling $500,000, and cannot pay more just because he believes or knows that the money supply will rise by 10,000% the next day, we instead have someone with zero cash and $30 billion in wealth being constrained from paying more than $30 billion in cash for anything, and cannot pay more just because he believes or knows that the money supply will rise by 10,000% the next day.
Even if Bill Gates was told that the money supply and prices would rise by 10,000% the next day, he would still be constrained to paying a maximum of no more than $30 billion, because that is what his wealth is worth in dollars given the current money supply.
“You don’t need money to buy homes, you need wealth. If the home is expected to be worth much more in the future, that will make it worth more right now.”
Home builders and sellers don’t accept “wealth” in exchange for the homes they sell. They accept dollars. You need to convert any wealth you have into dollars before you can buy a home. And the dollars you can fetch for your wealth will be a function of today’s money supply, not tomorrow’s money supply. It’s why houses were prices at $30,000 in the 1960s, and $300,000 today. It’s because there’s more money and spending. Even if people in the 1960s knew that the money supply and prices in 2012 would be 10 times higher, they would still pay $30,000 for homes in the 1960s, because prices at that time were, just like today, constrained to the money supply of the time.
You do adhere to the quantity theory of money, right? Aggregate demand and aggregate prices are primarily a function of the money supply that exists, not what might exist at some time in the future.
Suppose that Bernanke made a credible announcement that he will raise the money supply by 10,000% tomorrow. It doesn’t matter if you say that the current values of assets “should rise” today. What matters is what current prices can actually be paid today. That is where value is ultimately derived. From current, actual, relative prices. The prices that can be paid today cannot be 10,000% higher just because people expect that is what future inflation will be. Current prices will be capped by the current money supply, not future money supply. Sure, once Bernanke does inflate the currency, and people actually have more money in their pockets, THEN actual market prices can rise, because then people actually have the additional money to pay the higher prices.
22. January 2012 at 14:58
Scott,
“No, when I use MV=PY, I mean the base, and base velocity. base velocity is NGDP/base.”
But for the broader aggregates (i.e. money actually used in the transactions that make up NGDP) there is a clear relationship to PY. Unless you can say something about the mechanics by which the base impacts NGDP, it’s just an accounting identity. It’s like saying BV=PY, where B equals the price of bananas and V equals PY divided by the price of bananas.
I’m probably being dense here!
22. January 2012 at 16:48
Dtoh,
There is a relationship between the base and PY, though it’s an even sketchier one than the broader aggregates and NGDP. I like to think of it this way-
Base money > asset prices less base money > broad money > NGDP.
There is also-
Base money > asset prices > interest rates > velocity > NGDP.
– but the effect of base money on the interest rate impact on velocity seems to be highly unreliable and often transistory e.g. very high interest rates in Britain in 1980-1981 didn’t mitigate velocity’s plummetting dramatically as the value of the pound shot up.
22. January 2012 at 16:50
(Personally I prefer to think about the central bank changing the broader aggregates through changing asset prices and interest rates, rather than bringing base money into the picture. I think that money is only a useful concept when central banks stop targeting interest rates or if you really want to go into the mechanics of central banking.)
22. January 2012 at 21:40
Scott, so if I purchase machine tools, robots, motors, a pre-fab building, office equipment, desks, etc. and bring these together and begin cranking out widgets, then I haven’t invested anything, because almost the whole of project involved the purchase of goods constructed by someone else?
How much damage to common sense do we have to do to bring reality in line with the demands of Keynes’ other-worldly construct?
“Greg, You said;
“If any given firm which before used annually to buy a certain amount of machines in replacement of old ones decides to produce these machines in its own factories, the amount of money used to buy investment goods will decrease but the total amount of investment will not change.”
No, investment is not inter-firm purchases of capital goods, it’s funds spent on construction of capital goods.”
23. January 2012 at 08:46
Peden,
I know there is a relationship there, but I think the way Scott is using it, it’s just an identity, because if if you assume that velocity is a real concept (and not just a plug number to equate M to PY), then it’s trivial to show that Base x Base Velocity does not always equal PY thus disproving Scott’s model.
23. January 2012 at 08:53
dtoh,
Ah, I think I see what you mean.
Though if base velocity is defined as PY divided by the base, then there is a tautology. However, the entire concept of base velocity is dubious, if defined that way. I want to define base velocity as PT / Mb (since most of the Mb exists solely in the financial system at the central bank) which is a very different number from PY / Mb.
23. January 2012 at 09:17
If you are talking about Base money, then MV=PY (or PT)is either
A) An identity and meaningless, or
B) Wrong
23. January 2012 at 15:07
So to an extent, I buy this, but I think there are important cases where thinking of a discrete quantity of funds “sloshing” around is particularly useful.
There’s got to be a gold-standard safe asset. In the case of the most recent financial crisis, T-Bills, Gilts, and Gold have essentially served in that capacity. But remember the record-high oil prices that occurred in June/July 2008 right as NGDP expectations were leaning off a cliff and rapid disinvestment was occurring in the housing market? The only way that you can explain the persistence of that elevated level of an asset price fundamentally dependent on growth is that people hadn’t realized the future path of NGDP (a bad sign for futures targeting) or the bucket of money theory holds some water.
24. January 2012 at 06:25
Major freedom, You said;
“If you bank loans you $100, then you can start to write checks and make payments to merchants off that loan.”
You are confusing loans with DDs. Suppose I have a $100 saving bond from the US government, so it’s very safe, no default risk etc. I take that $100 savings bond to the store, what can I buy?
On the 10,000% rise in M, I’d expect V to rise immediately, in anticipation of the future rise in M.
dtoh, If the $ price of bananas falls in half because of a big banana crop, the NGDP in banana terms doubles. I think everyone agrees on that. The mechanism is that in this case the medium of account (bananas) has lost half it’s value, so if NGDP is dollar terms is constant, then NGDP in banana terms will double.
So there’s no doubt that changes in the medium of account can rapidly change NGDP. But that’s not interesting, because we don’t actually measure NGDP in banana terms. The monetary base is the medium of account in the US economy. So changes in the base will usually affect NGDP by affecting the value of base money. (This is more debatable at the zero bound.) The mechanism, at least during normal times, is the hot potato effect. People have a limited demand for cash. If the public prefers to hold one week’s income in the form of cash, then NGDP will always be 52 times larger than cash holdings. Double cash holdings, and you double NGDP. On the other hand their preference for cash (and bank reserves may change, and then it gets more complicated–like right now.
Greg, You said;
“Scott, so if I purchase machine tools, robots, motors, a pre-fab building, office equipment, desks, etc. and bring these together and begin cranking out widgets, then I haven’t invested anything, because almost the whole of project involved the purchase of goods constructed by someone else?”
No, if the way you organize those tools makes the collection of tools more valuable than they are individually in separate locations, then investment has occurred, If not, then not.
dtoh, MV=PY is definitely an identity, no matter what definition of M you use. V is simply NGDP/M, nothing more. It should not even be called “velocity.”
Shocking. No, the high level of oil prices was associated with a booming world economy, as soon as the world economy began to decline sharply, oil prices plunged. In any case, NGDP futures targeting is not built on the assumption that the public is particularly good at predicting the future. But I think they are no worse than the experts. Market saw the plunging NGDP problem long before the Fed, which was still in denial at their September 2008 meeting.
24. January 2012 at 07:41
ssumner:
“If you bank loans you $100, then you can start to write checks and make payments to merchants off that loan.”
“You are confusing loans with DDs. Suppose I have a $100 saving bond from the US government, so it’s very safe, no default risk etc. I take that $100 savings bond to the store, what can I buy?”
You are confusing GIVING a loan with RECEIVING a loan.
If you RECEIVE a loan, which was the context under discussion of a bank loaning you $100, then you get legal tender balance from which you can write checks off and buy goods. The lender gets the “bond” that isn’t legal tender.
If a bank loans you $100, then you don’t get bond. The bank gets a “bond.” You get legal tender that counts towards M3, and which you can use to buy goods such as houses.
My argument is that the banking system’s expansion of “soft” money credit, founded upon the Fed’s expansion of “hard” money in the banking system, is what financed the housing boom. The bonds the banks created out of nothing is on their end. The money they created is on the borrower’s end, and it is the money on the borrower’s end that financed the housing boom.
Saying I am confusing credit with money (which isn’t the case, since credit money counts towards M3), then saying I am confusing loans with DDs (which isn’t the case because I am talking about the newly created money that is loaned to the borrower, not the bond being created on the bank’s side), is quite honestly a baffling set of responses I can’t wrap my head around.
“On the 10,000% rise in M, I’d expect V to rise immediately, in anticipation of the future rise in M.”
V cannot rise unless there is an equivalent rise in the speed at which goods and services are produced and sold. However, more goods and services being produced and sold will reduce prices, not raise prices. So if V increases, given today’s M, then the same quantity of money is being turned over more often on the basis of a higher rate of productivity. That can’t raise prices today. Given a fixed M, prices can rise only with a lower demand for holding cash (and/or reduced real goods supply).
V is really just a fudge factor that enables the tautological MV=PT equation to be in equality. The only way to calculate V is to use V=PT/M. V is not an independent variable that can be defined apart from the other variables. V cannot be observed in itself. To say V rises is to really say PT/M rises. But if we’re trying to explain a rise in P, we cannot say “P rises because V rises”, since V is defined by taking into account P.
If I spend in a certain day $100 for food, and I had an average cash balance (or M) for that day of $1000, then by definition, V equals 1/10. If you want to explain why I paid $100, then you can’t point to V=1/10. I had an average quantity of money in my cash balance of $1000, each dollar turned over on the average of 1/10 of the time, and so I spent $100 in this period. OK, but it is absurd to take a quantity and give it a place in an equation unless it can be defined independently of the other terms in the equation. Since V cannot be defined independently of the other terms, it means it is a fudge factor.
In the aggregate, aggregate P is determined by aggregate demand and aggregate supply. If one expects that tomorrow’s P or tomorrow’s aggregate demand will be higher, then today’s P cannot rise unless there is more demand (or less supply) today. V by itself cannot raise prices. The only way there can be a rise in the aggregate price level is if there is a decline in the demand for money, or a rise in the quantity of money.
The MV=PT equation just says that total money received in a transaction is equal to the total money given up in a transaction. It’s essentially worthless.
Expecting a higher price level tomorrow can only increase prices today by way of those expectations generating a decline in the demand for money today, or a rise in the money supply today (and/or a reduction in real supply). But, and this is my point, if future price inflation expectations are greater than what would be possible to actually accommodate today through decreased cash balances, say 10,000% inflation expectations, to actually make today’s prices match tomorrow’s prices, then the money supply will be a binding constraint and prices will not rise any further.
Everyone is constrained to the cash they have or can get. If I have $500,000 in cash, or wealth, and I expected house prices to rise from $500,000 to $1 million tomorrow, I cannot raise my demand for houses to $1 million today. No, you can’t just change the example as say “Bill Gates the billionaire could”, because that will just mean we have to change the numbers while keeping the principle. So for Bill Gates, who has $30 billion in wealth, if he expected prices to double tomorrow, then he won’t be able to pay $60 billion for goods today. He can only pay a maximum of $30 billion for goods.
Since everyone is cash constrained today to the money supply today, it will be impossible for people to raise aggregate prices today beyond what they are able to pay.
24. January 2012 at 09:23
Scott,
You say, “So changes in the base will usually affect NGDP by affecting the value of base money. (This is more debatable at the zero bound.) The mechanism, at least during normal times, is the hot potato effect. …..Double cash holdings, and you double NGDP. On the other hand their preference for cash (and bank reserves may change, and then it gets more complicated-like right now.”
So in effect, you are saying is that all we can say about money in normal times is “If the Base increases then sometimes NGDP may increase,” but in downturns “It’s complicated and we can’t say anything.”
This seems a wholly unsatisfactory explanation of monetary policy. I don’t think you can make a convincing argument for level targeting NGDP, unless you have a better model to explain how the Fed can influence NGDP by manipulating M. Simply citing expectations is insufficient because expectations will change unless the market believes there is actually a bullet in the gun.
I also think that statements like “Money and saving affect the business cycle (if at all), by affecting either M or V” don’t make sense or are meaningless when V is simply an arithmetic plug to equate M to PY. Or can you actually calculate V independently by measuring actual actions such as the exchange of currency in payment for goods or services.
26. January 2012 at 10:58
Major Freedom, I’m not talking about giving or receiving loans, but rather loans as an asset. I don’t follow your comments on velocity.
dtoh, You said;
“So in effect, you are saying is that all we can say about money in normal times is “If the Base increases then sometimes NGDP may increase,” but in downturns “It’s complicated and we can’t say anything.””
No, I could write a 400 page book on all the things we can say about the base, but I can’t do that in each answer. I have to try to summarize the main points.
No, V can’t be calculated independently. It’s true that equation is a tautology, I pull it out because many believe that monetary policy only affects M. It also affects V and I sometimes like to remind people of that fact.
There is an enormous literature on the transmission mechanisms of monetary policy. Mishkin’s textbooks discusses no less than 10 transmission mechanisms, and I’ve discussed many in this blog.
If you google my “short course on monetary economics,” I discuss some ways that money affects NGDP.
27. January 2012 at 09:40
ssumner:
“Major Freedom, I’m not talking about giving or receiving loans, but rather loans as an asset.”
The person who owns the “asset” is the person giving the loan. The person who owns the money is the person receiving the loan.
The people who receive the loans used those loans to bid up the prices of houses. They got those loans from the banking system. The banking system creates loans out of thin air that are not backed by prior real savings.
“I don’t follow your comments on velocity.”
Velocity can only rise to the extent that the speed of production and delivery of goods and services can rise. Velocity is a fudge factor that equates the “goods multiplied by prices” side with the “money supply multiplied by fudge factor” side.
27. January 2012 at 09:46
“Major Freedom, No, credit is not money. If my local banker loans me $100, he can’t take that loan to Walmart and buy something.”
No, but YOU can, and that’s the point. The more credit banks issue ex nihilo, the more money borrowers like home buyers have to buy houses, and the higher home prices will go.
30. January 2012 at 16:30
Scott,
So I read Mishkin. Thanks for the suggestion. A couple of thoughts.
1. It’s not a very quantitative explanation, but I suppose that’s to be expected in a textbook for a summary course.
2. Even if you could mathematically formulate the mechanism, you couldn’t solve it (just like a multi-body orbital equation) so from a policy perspective, it’s not very useful in arguing for a specific policy recommendation.
3. It doesn’t really address the hot potato mechanism.
4. It addresses the mechanism from the perspective of the broader aggregates (which I think I already understood), whereas your argument is more focused on the base where to me anyway the mechanism is less clear or at least less tightly linked.
31. January 2012 at 14:10
Major Freedom, You said;
“Velocity can only rise to the extent that the speed of production and delivery of goods and services can rise.”
Not at all, if they don’t rise you get inflation.
I think we are talking past each other on loans. I thought you were claiming loans were a medium of exchange. You seem to be refering to DDs, not loans.
dtoh, Obviously I differ in certain respects, but the point is that there are many transmission mechanisms. I focus on the effect on asset prices, and also the expected future hot potato effect.
31. January 2012 at 16:46
Scott,
I’m not sure I can make a credible argument (yet) but I think the credit mechanism (closely related to asset prices) is very important, which is why I keep harping on about the Fed directly regulating bank asset/equity ratios (both the minimum and maximum).