Problem Set 7
Principles of Economics, fall 2009
1. How large a change in government spending is needed to close a recessionary gap of 200 if the multiplier is 4? If it is 1.5?
2. Based on the table below, sketch an aggregate demand curve and an aggregate supply curve.
a. What is the equilibrium price level and output?
b. If aggregate demand shifts out, what happens to equilibrium output?
c. If aggregate demand shifts in, what happens to equilibrium output?
d. What role do changes in aggregate demand play in the economy with these aggregate supply and demand curves?
Price Aggregate supply Aggregate demand
90 3000 3500
95 3000 3000
100 3000 2500
105 3000 2200
110 3000 2000
3. Economists with a Keynesian orientation sometimes argue that central banks that practice inflation targeting don't care about unemployment. Explain this argument in a Keynesian AS-AD framework. Compose a rejoinder in a neoclassical AS-AD framework.
4. How can deposit insurance discourage a bank run?
5. When a bank is rumored to be in trouble, why won't other banks lend it money? Why is a lender of last resort needed? Iceland has a population about half that of Staten Island. Explain why Icelandic banks that operated throughout Europe might get into trouble.
6. Suppose the only two countries in the world are the US and China. Everything will be discussed from the US standpoint. Let
e = exchange rate = yuan/$
("Yuan"is the colloquial name for the Chinese currency.) If e goes up, the dollar appreciates.
US demand for imports from China is
M = 5 + e/4.
All supplies and demands are denominated in hundreds of billions of dollars.
Chinese demand for exports from the US is
X = 10 - e/4.
Demand by private parties in China for US capital like US stocks, New York condos, and US treasury bills is
X capital = 5 - e/4.
Demand by Americans for Chinese capital like Chinese stocks, Shanghai condos and Shenzhen municipal bonds is
M capital = 5 + e/4.
a. If the yuan is allowed to float freely, what is the exchange rate?
b. The Chinese government wants to maintain an exchange rate of 7 yuan per dollar. What should it tell the Renmin Bank (the central bank of China) to do? Give an answer that includes a quantity.
c. What is the US trade deficit with China (when China maintains an exchange rate of 7 yuan per dollar)?
d. In the global panic of 2008, investors everywhere seek safety, and US treasury bills are the safest investment. Thus the demand by private parties in China for US capital stock becomes
X capital = 5.5 - e/4.
Similarly demand by Americans for Chinese capital falls to
M capital = 4.5 + e/4.
The Chinese government keeps the exchange rate at 7 and nothing else changes. How does the Renmin Bank respond? (Once again, an answer with a quantity as well as a direction.)
e. What is the US trade deficit now?
f. Suppose the marginal propensity to consume in the US is 0.6. How does aggregate demand change as a result of the popularity of US treasury bills?
g. What would happen in the US if China allowed the dollar to depreciate?
h. What would happen in China if China allowed the dollar to depreciate?
7. The only export of the Republic of Alterior Terra (RAT) is cheese. It's a distinctive cheese, and studies have shown that the price elasticity of demand for this cheese in the export market is -0.7. RAT is considering whether to devalue its currency, the Mouse, by 10%.
a. If RAT devalues, by what percentage will the quantity of cheese it exports change? Up or down?
b. How does the demand for Mouses change? Up or down?
Money and Monetary Policy Notes—November 2009
Last week we saw that sticky prices and wages in the short run could keep a market economy from Pareto optimality by causing unemployment. In the short-run the mechanics of price adjustment didn't work out. We need one more piece to be able to talk about the short run intelligently—money and banking. Topic for today. Everything today will be dealing with a "closed economy" this week—no foreign sector. Kind of old-fashioned. Soon we'll expand to the international financial system.
Money is the technology we now use to carry out complex transactions. We don't need it. One alternative to money is barter: trading goods for goods directly. But barter is informationally demanding: we need n(n-1)/2 different prices to barter n different goods, and either double coincidence of wants or a giant scoreboard. Like hub-and-spoke airline scheduling—why everything goes through Denver. Double coincidence of wants. Given the information technology of the 20th century, money was the best way of storing and retrieving this information. But it may not always be. Ways of transacting business are changing very rapidly. I expect that most of this lecture will be a useless historical artifact a decade or two from now.
So money is a tool for making trades. The characteristic of money that makes it a useful tool is liquidity. (Alternatively, anything that is liquid is money.)
We use many other tools for making trades. Telephone, internet, lawyers. Language: I can't order 'vegetarian pad Thai' unless I know what it means and so does the person I'm talking to. Time and calendar: 'next Saturday' has to have an agreed meaning. Standards, too: AA batteries, 2 x 4 boards, Word documents, screws, floor slipperiness.
Many of these tools are conventions. We could speak Esperanto or Hungarian instead of English, use sidereal time or the Islamic calendar, and do just as well. We could use blue money as well as green, pennies or quarters as well as dollars. Everything real would be the same: the message would be different, but the meaning would be the same.
But predictability is important. You don't want to get a hamburger or a Lexus when you order vegetarian pad Thai; you don't want it to take 2 seconds or 10 years when they say 20 minutes. Predictability doesn't have to be perfect: you can be a little surprised at what your vegetarian pad Thai looks like. Stasis is not required: if the meaning changes predictably that's ok. "A nice day in the winter" is not the same as "a nice day in the summer," but we understand what's meant by both.
The cost of inflation—unpredictable inflation—is that it makes the tool less valuable. If you're the keeper of the tool—a central banker—you don't want the tool you keep to become less valuable (even if the social cost is really not that great).
If we're in the long-run world with fully flexible prices and Pareto optimality, that's all that needs to be said about money. Money can be boring (like a calendar) or it can be pernicious by being unpredictable; there is no other role in this model. Money is also conventional in this model: it doesn't matter what it is and how prices are quoted as long as it's boring and predictable. We care about the meaning, not the message. That's why we didn't talk about money in micro or long run macro.
But short-run fixed price macro is different. Why? The inflexible prices and wages that are the root of the problem here are nominal wages and prices. The message is fixed—but the meaning isn't. And the meaning is wrong. So it's possible that if you manipulate money you manipulate the meaning: for example, lower wages without workers knowing it. This is the deep reason why money matters in short-run macro.
2. Interest rate
The next new wrinkle is " the" interest rate. We said last week that some wages and prices were sticky, not all. Interest rates aren't sticky. We're concerned with the price of future money—expressed in the nominal interest rate—and the price of future consumption—expressed in the real interest rate.
Digression or reminder. The nominal interest rate is the amount of extra money you get next year for giving up $1 this year. That's what's usually quoted. The real interest rate is the amount of extra output you get next year for giving up a dollar's worth of output this year. If the nominal rate of interest is N and the rate of inflation is I, the real rate of interest is N-I (approximately). If you give up a dollar's worth of output this year, you get (1+N) dollars next year, which will buy you (1+N)/(1+I) units of output next year, and (1+N)/(1+I) is approximately 1 + (N-I).
Let's go back to last week. We talked about investment in the income-expenditure, Keynesian-cross world. Investment depends on the real interest rate, as well as on animal spirits. Some projects worth doing at 2% interest are not worth doing at 10% interest. Savings, too, often depend on the real interest rate—for instance, in the days when people refinanced mortgages every 5 minutes (the lower the interest rate, the more you took out of the refinancing and the lower your NET savings were). When the real interest rate goes down, consumption and investment both go up, holding income constant. Multiplier works on this.
So we need now is a theory of the real interest rate in order to find the short-run equilibrium.
Here's where money comes in. The interest rate that the market sets is the nominal interest rate. With some rare exceptions, you can't trade future and present consumption directly; only future and present money. That's a feature of the 20th century; don't know if it will still hold through all of the 21st. In the short run, inflation expectations are fixed, so nominal determines real.
What determines the nominal interest rate in the short run? Supply and demand, with instant adjustment. What's different from a normal market is that the Federal Reserve open market committee is the big player.
3. Federal Reserve
You learned a lot about the Federal Reserve in section, and I don't need to repeat it here.
One fact that doesn't show up in text books: The Fed's money is actually taxpayer money. Taxpayers are the residual claimants on Fed profits and losses (the Fed has not had a loss since 1915). The Fed uses the income it gets from its investing activities to cover its operating expenses, and whatever is left over goes to the Treasury. In 2008, the Fed had investing income of $39 billion, and remitted about $35 billion to the Treasury. (Since in normal times most of the investing income comes as interest on federal debt, this is really a wash, except for the operating costs of the Fed.)
The textbook picture emphasizes the role of the Federal Reserve Open Market Committee (FOMC). This is a committee of some of the Board of Governors and some of the regional bank presidents. It meets about every seven weeks and decides what it wants the 'federal funds rate' to be. This is the rate at which banks lend balances to each other overnight. They need to lend balances to make sure they maintain their required reserves. The federal funds rate in recent days has been about 0.13%. Since December 16, 2008, the FOMC has set the rate as a range from 0-0.25%. It can't get lower. Before the crisis, the rate was about 5-6%.
The New York Fed carries out decisions that the FOMC makes.
It usually operates in the T-bill market. T-bills are short-term bonds of the US governments (T stands for Treasury). The Fed buys and sells T-bills to get the interest rate it wants. It can buy any amount it wants because it can print money to do so. It can sell any amount it wants, because it holds an awful lot. Other people and institutions are in the market for T-bills for all sorts of reasons, but the Fed offsets whatever they do to get the interest rate it wants. (Actually the Fed doesn't mainly buy and sell T-bills any more. Instead it enters into repurchase agreements for a few days. That way its position automatically reverses quickly.)
When the Fed buys T-bills, money supply goes up, because it prints the money that uses (actually, it uses checks). When it sells T-bills, money supply goes down.
4. How it matters—IN NORMAL TIMES
In normal times, the ability to set short term T-bill is important because it influences investment and savings and the exchange rate (more about this later). But investment depends on long-run interest rates and the stock market; savings mainly on long run (refinancing). How does the influence spread?
For other short-term investments, T-bills are a good substitute in normal times. So other short term interest rates usually move in sync with the T-bill rate. For long term rates, the influence is more indirect. Partly current short term rates say something about short term rates in the near future, and a long term is just a combination of near futures. If I can get 10% now on 1-month notes, then to get me to buy a 10-year notes I need an interest rate that will compensate me for the 10% I expect to get not just this month, but at least for a few months after that.
Short-term rates also matter for construction. If you want to build an office building, you will get a short-term loan to pay for construction. Once it's complete, you will then go out into the market again to "take out" the construction loan with a long term loan. Why not just get a long term loan? Because without a constructed building, you don't have good collateral for the long term loan.
There is also an effect through the behavior of banks. This is a direct liquidity effect. The textbooks don't emphasize it, but it's always there; it's just been more important since the crisis began. Consider a bank getting ready to make a big loan. If the federal funds rate is high, it may decide to wait until it has enough reserves in order to make that loan without getting into trouble. And it may worry about that loan a lot, because it will deplete its reserves for as long as it's outstanding and make further lending more difficult. But if the federal funds rate is low, getting caught short on reserves is no problem: it can easily borrow overnight at a very low interest rate to take care of any problem that might occur. So a low federal funds rate normally makes banks more eager to lend, since they don't have to worry as much about the liquidity consequences.
Opportunities in other countries and capital flows to and from other countries, if they're big enough, can offset part of what the fed does on long term rates. Will have to talk about international money system to understand this.
When investment increases , aggregate demand goes up. Keynesian cross. AS-AD. This raises prices and real GDP. If in a recession, most of the gain is in real GDP. If not, most of the gain is in prices—depending on the slope of the SRAS curve.
Time lag is considerable: a year maybe. Cutting interest rates often does a good job with a moderate recession.
5. What about now?
Standard monetary policy is a band-aid; it doesn't do much good when you're having a heart attack.
The federal funds rate is very low now, but in a deep slump it's not enough. It's also not enough to deal with financial crisis.
To understand why, first we need to review what the financial crisis was about, and then think about what's going on now. Then I'll talk a little about what monetary policy means in the Great Recession.
a. The financial crisis
When we talked about financial intermediaries, we saw a lot of this. In 2007, regular banks held many interesting assets, and regular banks were not the only financial intermediaries.
For regular banks the story was pretty simple. They had a lot of assets which were not worth what they thought they were worth—FF stock, RMBS, in particular. If you valued it at market prices, they did not have sufficient reserves: in fact, they may not have had any capital at all. Moreover, nobody was sure who was in trouble or who would go next, and so they were reluctant to lend to each other. Because they might not have capital and because they would have no way of paying off claims against them, they were very reluctant to lend. If they had cash, they wanted to keep it to protect themselves.
Investment banks were in a similar position, but they did not have formal reserve requirements the way regular banks did. But they still had to be prepared for runs. Same for money market funds.
The repo market is particularly interesting. Remember that the investment banks and regular banks often used RMBS and CDO as collateral because these were liquid. They suddenly became illiquid. This sort of lending became greatly restricted.
All of this lending dried up. Remember that money is a tool, and anything that is liquid is the same kind of tool. The quantity of LIQUID SECURITIES took a huge hit in 2008, and this made it much harder to do the jobs (like arranging construction financing) that the tool was being used for. Like a blackout that cuts off electricity to Manhattan. When you think of 'money' as any liquid security, which is the right way, then the amount of money that was circulating fell dramatically. (Nothing about money says it has to be issued by governments or government agencies, and for most of US history it wasn't.) You can think of one of the things that happened last year as a huge reduction in the 'money supply', and just work back all the implications in the AS-AD diagram.
b. Where we are now
The result of all of this, as I said last time, is that you're coming back from the gym with a broken arm, not muscle soreness. That adds new problems for standard monetary policy.
Money alone is not enough to get banks to loan: they have to see profitable opportunities, and they have to be secure that they have enough capital, and they have to think that lending is the most profitable thing to do with their money. For many banks we're not there yet.
Driving down short-term interest rates doesn't always spur investment; nor does providing liquidity.
c. New style monetary policy
So this year the Fed (and the Treasury) have rewritten monetary policy. They have decided that buying short-term T-bills is not enough to deal with the recession we are facing. They are addressing liquidity issues as well as lending issues. What are they doing?
Term auction facility. Traditionally, the Fed has made short-term collateralized loans to banks to help them get enough reserves, and charged them interest for it. This is called the discount window. But for several decades, this has been used very little, because some stigma was attached to it. Under the TAF, the Fed auctions off 28-day loans, which banks can use to meet their reserve requirements. The auction (and how much the Fed is offering) determines the interest rate, which is higher than the federal funds rate. Another way of making banks liquid, but without adverse selection: you don't have to identify yourself as needy to participate in the auction.
Term securities lending facility. The Fed lends treasury securities to 'primary dealers'—the 20 biggest firms it usually deals with. It takes stuff that would not otherwise be liquid, like investment grade RMBS, as collateral. But there is a haircut—the collateral is valued more than the loan. And if the collateral loses value, the Fed can ask for substitute collateral. This isn't new money, but it improves the liquidity of the primary dealers.
Primary dealer credit facility. Many of the primary dealers were not part of the Federal Reserve system (eg, at the beginning last May, the investment banks), and so were not allowed to borrow at the discount window. This allows them to borrow cash at the discount window—with collateral, which does not have to be stellar.
Capital positions in banks. The Treasury started off last year by lending huge sums to many banks, and making it known that they would be willing to lend more. In return for these injections of capital, they took various stakes in the business. This reduced the pressure for runs on these intermediaries, and kept many of them in business. It has meant that they are still around to lend. The downside is moral hazard, which we discussed earlier.
Guaranteeing money market funds. Ditto. By the Treasury
Long term bonds. The Fed is now buying long-term treasury bonds, as well as short term T-bills. The idea is to reduce long-term interest rates directly, and so get investment going directly.
FF bonds. The Fed is buying Fannie Mae and Freddie Mac bonds and RMBS (the goal is $1.45 T by the spring). This is also to reduce long-term interest rates. This has its most direct effect on the housing market, which I'll talk about in a day or two. It's not neutral what the Fed buys. (Especially if you're thinking about issuing private label RMBS.)
Commercial paper. These are short-term loans by highly-rated corporations and banks. Normally money market funds and banks would buy this paper. The Fed is buying almost directly, and so funding banks to encourage them to lend, and making it easier for nonbank corporations to keep operating. ("Almost" because the law doesn't give the Fed the authority to do this. Instead, it has set up a separate corporation -a 'special purpose vehicle'—and is lending to that corporation, which in turn buys the CP). This is the Fed's second shot in this market—originally, it offered non-recourse loans to banks that bought the relevant CP. (Non-recourse means that if the ABS lost value, the bank would be liable only for the value of the CP). There is a separate fund set up especially to buy CP from money market funds—so they don't have to worry about its liquidity. Again this is not neutral: it helps certain classes of borrowers—those big enough to be able to issue CP—and not others.
Term asset-backed securities loan facility. The Fed 'supports' asset-backed securities (ABS) issued by banks, when the assets are auto loans, credit card loans, student loans, equipment loans, floorplan loans, insurance premium finance loans, loans guaranteed by the Small Business Administration, commercial mortgage loans, and residential mortgage servicing advances. Issuance of non-mortgage asset-backed securities fell from $902 b in 2007 to $5 b in the fourth quarter of 2008.
"Support" works like this: The Fed makes non-recourse loans to certain big guys (primary dealers or subsidiaries) to purchase recently issued ABS with high credit ratings, with the ABS as collateral. The collateral exceeds the value of the loan, to give the Fed some protection. The loans are pretty long (up to three years). If the loans are not repaid, the Treasury will bear the first $20 billion in losses, but this doesn't really matter, since everything is coming out of the taxpayer's pockets eventually. This is not neutral either, in that it helps certain kinds of activities, not others—like software or R&D.
Paying interest on bank reserves. Banks are required to keep reserves with the Fed, as you learned in section, and up until this year, these reserves did not pay interest. At the end of 2008, Congress authorized the Fed to pay interest—not just on excess reserves but on required reserves. The interest on excess reserves encourages banks to be sound, which is good, but discourages lending. It gives the Fed another tool for controlling bank behavior, which may be useful in fighting inflation. But the interest on required reserves is a pure transfer from taxpayers to banks, since it can't alter their behavior.
Bear Stearns. When Bear Stearns was sold to JPMorgan in March 2008, the Fed (acting through a special purpose vehicle) bought about $30b in dicy securities from Bear, to make the deal more attractive to JPMorgan. It may or may not make money on this investment; probably not.
AIG. Too complex for words. Basically, it has done the same thing with $43 billion in AIG assets, as well as extending a direct loan of similar size.
Citigroup and Bank of America. In both cases, the Fed and the Treasury guaranteed pools of assets worth hundreds of billions. The banks will take the first defined losses on the pools, then the Fed and the Treasury will take 90% of the remaining losses. For this insurance, the banks will pay the Treasury a fee in cash and in stock.
d. The bottom line
The Fed is trying to move the AD curve out. It is trying to find new and different ways of doing this, because the old ways won't work in the current situation. It is sterilizing some of these new interventions: it is reducing its net purchases of short-term T-bills in order to keep federal money supply from expanding too quickly. Thus new instruments are supplanting the old instruments.
I don't expect you to learn all of these fancy facilities. I can barely keep track of them when I have documents open. Many of them will be gone by the time you graduate from college. But I don't expect them all to disappear.
Remember that money is essentially an information technology tool. As information technology evolves, money evolves. As money evolves, monetary policy should evolve too. And you can watch it changing right before your eyes.
Addendum on 'printing money'
In my discussions with people after class last night, I realized that 'just printing money' has become some sort of a pejorative expression, as in, "Isn't it horrible that the Fed is just printing money?" This is a very unfortunate way of looking at things.
First, if there is going to be money, somebody has to print it. As your parents told you long ago, money doesn't grow on trees.
Second, printing money is less interesting than the methods the Fed uses for distributing money. After all, if the Fed just printed a whole bunch of dollar bills and stuffed them in its vault, it would not make any difference to anyone except the Fed's janitors. Most of the lecture was about the ways the Fed distributes money (buying securities or lending money) rather than printing it.
Third, money is a tool; it's not a sacred object to be venerated. How much of this tool we need at any particular time is a policy decision that sometimes the Fed makes right and sometimes the Fed makes wrong. You can criticize any particular decision or any pattern of decisions, but it's not sensible to sneer at any decision-making whatsoever.
Finally, it you look at money as liquidity, the problem was created when AIG, Lehman Brothers, Morgan Stanley, Citibank, Goldman Sachs et al were essentially all printing money, and then that money disappeared. The Fed printing a lot of money to replace that money is not that inappropriate. The problem is that nothing is stopping GS, Morgan Stanley, JPMorgan, and whoever is left standing to print more money in the future. (If these guys are too big to fail, why should I worry when I get an IOU from them? And how different is that from not worrying when I get an IOU from the Fed?). In that sense, the problem is that we may end up with too many entities "printing money".
Please see the attached file.
In the numerical question on exchange rates there were some issues. The way ...
The money and monetary policy notes for principles of economics is determined.