Economics816 Midterm

Prof.Bryan Caplan


Mean: 86        Median:86


Part 1: True, False,and Explain

(10 points each - 3 for the right answer, and7 for the explanation)

State whether each of the following ninepropositions is true or false.  Using2-3 sentences AND/OR equations, explain your answer.


1.  True,False, and Explain: Both New Keynesian and RBC theorists accept the mainconclusions of Friedman's "The Role of Monetary Policy."


FALSE.  Friedman and the NKs emphasize the short-runimpact of money on real variables.  RBCtheorists model the economy without appealing to nominal --> real causalmechanisms.



2. True, False, and Explain: Just aseconomic theory cannot determine how much the quantity of wheat demandedwill increase if the price of wheat falls by 1% (ceteris paribus), it cannotdetermine how much the nominal quantity of money demanded will increaseif the price level increases by 1%.


FALSE.  In the standard model of money demand,people want real balances, not nominal balances.  Therefore, desired real balances do not depend on the price level,and desired nominal balances must be directly proportional to the price level.


(Severalpeople appealed to the MV=PY identity in answering this question; they gotpartial credit, but either way, there was a problem.  If they said FALSE, they had to add special assumptions on thebehavior of V and Y; if they said TRUE, they weren't taking the basic model ofmoney demand into account.  Withoutextra assumptions, MV=PY is an identity not a theory).



3. Someone invents a new theft-resistantwallet.  People buy the wallet iff E(T)³Pw,where E(T) is a person's expected fall in losses from theft, and Pwis the price of the wallet.


True, False, and Explain:This invention will not only permanently raise the rate of inflation; it willalso cause a discrete jump in the price level.


FALSE.  The invention causes a discrete FALL in theprice level; since people now feel safer carrying cash in their wallets, thereis a one-time increase in the demand for money.  This has no permanent effect on inflation though, since theproblem stipulates that everyone who benefits from the wallet gets one atonce.  If use of the wallet graduallybecame popular, there could have been an extended inflationary effect.


(Mostpeople totally ignored the connection between money demand and risk oftheft.  I don't know why).




4.  True,False, and Explain:  Populationgrowth is deflationary.


TRUE.  More people means more demand for realbalances; holding the money supply constant, the price level has to fall toaccommodate this demand.


(Ifyou appealed to the supply effects of population growth or MV=PY, you got mostof the points, but not all because the explanation wasn't completely convincing[what about velocity effects and demand effects?].  Several people said FALSE, arguing that population growthincreases demand.  Even if this werecorrect, what about supply?   And itactually isn't correct, because demand for money increases along with demandfor goods!)



5. Suppose that, as in Fremling and Lott (1996), people can make unbiasedestimates of the impact of any variable if they put it into their model.  Otherwise, they assume it has noeffect.  The true model of real output(y) is:




Where y* is the free-trade level of output,Protection is a 0-10 measure of the degree of protectionism (0=free trade,10=autarky), and e is a mean-zero randomvariable. 2 out of every 3 people realizes that the level of Protection somehowaffects real output.


True, False, and Explain:The average estimate says that y will fall by 4 if the level of Protectionrises from 3 to 5; but if random errors by the people with the correctmodel are large enough, a majority of people might still deny that greaterProtection has a negative impact on real output.


TRUE.  1/3 of the people omit the variable, so theydeny that Protection has a negative impact on real output.  The remaining estimates are centeredat 6.  But with large symmetric randomerrors, you could easily have, say, 1/3 of the people with the right modelestimating a coefficient of 0 or more (and another 1/3 estimating a coefficientof -12 or less!).  1/3 uninformed +2/3*1/3=55% denying the negative impact of Protection.


(Thepoint of this problem is just to show that the median citizen could bewrong about the sign of the coefficient in Fremling and Lott's model).


6.  True,False, and Explain:  Because theincome and substitution effects work in opposite directions, RBC models can neverpredict the direction of the impact of tax policy changes on labor supply.


FALSE.  A lump-sum tax has only an income effect, nosubstitution effect.  So RBC models canunambiguously predict an expansionary effect of increased lump-sum taxation.



7.  InRogerson's (1988) first example, there is a computational mistake.  Since , m=ln 2.5, and a=.5,, not .4.  Recallfurther that .


True, False, and Explain:  The fully insured equilibrium will set c= for everyone.


TRUE.  The fully insured equilibrium equalizesconsumption, and everyone has one unit of capital.  Therefore, total production is .  Since the totalpopulation is normalized to 1, everyone gets this exact amount of consumption.



8.  True,False, and Explain:  A centralfinding of ADP is that moving from moderate to high inflation could actuallypermanently reduce the unemployment rate.


FALSE.  ADP agree with the consensus view that highinflation has no long-term beneficial effect on unemployment.  What they claim is that moving from zero orlow inflation to moderate inflation could permanently reduce the unemploymentrate.



9.  True,False, and Explain:  If bothprices and nominal interest rates were rigid in the short-run, then no"transmission mechanism" for monetary policy discussed in class willoperate.


FALSE.  Numerous other transmission mechanisms werediscussed in class: credit rationing/ bank lending channels, and the realbalance channel being the most obvious.


Part 2:Short Answer

(20 points each)

In 4-6 sentences AND/OR equations, answer allthree of the following questions.


1. Consider a model where the price level is unresponsive to monetaryshocks for 1 period but completely flexible thereafter.  Monetary growth for all periods beforeperiod 1 has been 0%; from period 1 on, its rate of growth unexpectedly butpermanently shifts up to 1%.  Peoplehave rational expectations and realize that this permanent change has occurredduring period 1. m0=y0=0.


The IS and LM curves are given by:


IS: y=c-ar                    LM:m-p=b+hy


Determine p0, p1, p2,and p3.  Is there ever a"discrete jump" in the price level? Why or why not? (Hint: You do not need to solve more than a coupleequations to get the answer).


Tosolve for p0, just plug into the LM curve.  m0-p0=b+hy0, so -p0=b,so p0=-b.


Tosolve for p1, simply note that prices are fixed for 1 period, andthe new info arrives in period 1. Therefore, p1=p0=-b.


Tosolve for p2, note that the change in monetary policy is now fullyanticipated, so real output will be back to its normal level of 0.  Further note that money growth has beengoing on for 2 periods, so:


m2-p2=b.  Since money is expressed in logs, 2%continuous growth from 0 sets m2=2. Thus:




Tosolve for p3, just note the continuing growth of the money supply: p3=3-b


Onthe discrete jump:  Since the LM curvehas NO interest-sensitivity, the standard result that a permanent increase inthe rate of monetary growth leads to a discrete price jump does NOT hold.  Normally, people are reducing their holdingsin response to the higher inflation tax (higher nominal rates).  If money holdings are insensitive to nominalrates, this doesn't happen.


(Igave most of the points to people who said that there WAS a discrete jumpbetween periods 1 and 2.  To get fullpoints you would have needed to distinguish this price-rigidity related jumpfrom the standard discrete jump).


2.  In an unassigned article, Kevin Grier findsevidence that in Mexico, a presidential "political businesscycle" exists, but has no effect on real variables.  Use Caplan's model of rational irrationalityto suggest an explanation for both of these facts.


As apolitical actor (voter), the marginal private cost of irrationality about theeffect of macro policy is zero.  As aneconomic actor (worker, etc.), the marginal private cost of irrationality aboutthe effect of macro policy can be substantial. So: As voters, people may passionately favor candidates who try to pumpup the economy during election year; candidates respond by using expansionarypolicy as political business cycle theory predicts.  However, as workers people realize what is going on, and demandpay hikes, cost-of-living adjustments in their contracts, and so on.  The result is expansionary policy that failsto improve economic conditions and in fact makes inflation worse.



3. Which RBC mechanisms does the evidence presented in Bernanke andParkinson (1991) cast doubt upon? 


Bernankeand Parkinson (not Bernanke and Blinder, the article a few people discussedinstead!) challenges a key RBC mechanism: the sensitivity of labor supply tomulti-factor productivity shocks. B&P claim that a productivity explanation of the Great Depression ishighly implausible. Thus, if RBC works for the post-Depression era, thestandard RBC prediction of procyclical labor productivity should fail for theDepression era.  In fact, it doesnot.  B&P's evidence does notdirectly challenge the many other RBC mechanisms (like the interest-sensitivityof labor supply) that have been proposed.