end of chapter 5

5.5 HOW DOES THE IS-LM MODEL FIT THE FACTS?
GOODS AND IINANOA1 MARKETS. THE /S-LM MOD. I
chapter 5
Describing the adjustment process implied by all these sources ot dynamics is obviously compli¬cated. But the basic implication is straightforward—it will take some time for output to adjust to changes in fiscal policy or to changes in monetary policy. How much time? This question can only be answered by looking at the data and using econometrics. Figure 5.12 shows the results of such an econometric study, which uses data from the United States Irom I960 to 1990.


(a)
1.6 -I
1.6 - 1.2 - 0.8 - 0.4 - 0.0 -0.4 H -0.8 -1.2 4 -1.6
„ 1.2- з a 3 0.8 о
•S 0.4 H
м on -
™ 0.0

confidence band
V--
I
S -0.8
-1.2
-1.6
Effect of 1% increase in federal funds rate on retail sales
I I I I I I I I I 1 I I
4 8
Time (quarters)
(b)
1.6 - 1.2 - 0.8 - 0.4 -
0.0
-0.4 - -0.8 - -1.2 - -1.6
Effect of 1% increase in federal funds rate on output


ll l i i i M l l l l l
4 8
Time (quarters)
(c)
Effect of 1% increase in federal funds rate on employment
l I I I 1 I I I l-l l l
4 8
Time (quarters)
Figure 5.12 The empirical effects of an increase in the federal funds rate
(d)
Effect of 1% increase in federal funds rate on the unemployment rate
H
Effect of 1% increase in federal funds rate on the price level




0.15 0.12 0.09 0.06 0.03 0.00 -0.03 -0.06
1.6 1.2 0.8 0.4 0.0 -0.4 ■ -0.8 -1.2 -1.6
01 м с re x и
QJ M
2

I I I l-l I I I M I I I
4 8
Time (quarters)


In the short run. an increase in the federal funds rate leads to a decrease in output and to an increase in unemployment, but has little effect on the price level.
SOURCE: Lawrence Christians). Martin Eichenbaum and Charles Evans. 'The effects of monetary policy shocks: evidence from the flow of funds'. Re.viev, of Economics and Statistics. February 1996. vol. 78. no. I. pp 16-34. © 1996 by the President and Fellows of Harvard College and the Massachusetts Institute ofTechnology.
cb/ipter 5


The study focuses on the effects of changcs in the federal funds rate, the interest rate that is most directly affected by changcs in monetary policy. It traces the typical effects of such a change on several macroeconomic variables.
I nch panel in figure 5.12 represents the effects ot the change in the interest rate on a given variable. Each panel plots three lines. The solid line in the centre of a band gives the best estimate ol the elfect ol the change in the interest rate on the variable we look at in the panel. The two dashed lines and the tinted space between the dashed lines represent a confidence band, a band within which the tnie value of the effect lies with 60 per cent probability.
• figure 5.12, panel (a) shows the elfects ol an increase in the lederal lunds rate ol I percent on retail sales over time. The percentage change in retail sales is plotted on the vertical axis, and time, measured in quarters, on the horizontal axis.
Focusing on the best estimate—thc solid line—we see lhat the increase in the lederal lunds rate leads to a decline in retail sales. The largest decrease in retail sales, -0.0 per cent, is achieved alter live quarters.
• Figure 5.12, panel (Ы shows how lower sales lead lo lower output. In response to the decrease in sales, firms cut production, but by less initially than the decrease in sales. Put another way, lirms accumulate inventories lor some time. I he adjustment ot production is smoother and slower than the adjustment ol sales. The largest decrease, -0.7 percent, is reached after eight quarters. In other words, monetary policy works, but it works with long lags. It takes nearly two years for monetary policy to have its hill effect on production.
• Figure 5.12. panel (c) shows how lower output leads to lower employment: as firms cut production, they also cut employment. As with output, the decline in employment is slow and steady, reaching -0.5 per cent after eight quarters The decline in employment is rellected in an increase in thc unemployment rate, shown in panel (d).
• Figure 5.12, panel (ej looks at the behaviour ol the price level. Remember that one ol the assumptions ol the IS-I.M model is that the price level is given and so doesn't change in response to changes in demand. Panel el shows that this assumption isn't a had approximation ol reality in the short run. The price level is nearly unchanged for the lirst six quarters or so. Only after the lirst six quarters does the price level appear to decline. This gives a strong hint as to why the IS—LM model becomes less reliable as we look at the medium mn—in the medium run. we can no longer assume that the price level is given, and movements in the price level become important.
Figure 5.12 contains two lessons. First, it gives us a sense ol thc dynamic adjustment ol output and other variables to monetary policy. Second, and more fundamentally, it shows that what we observe in the economy is consistent with the implications ol the IS-LM model. This doesn't prove that the IS—LM model is the right model. It may be lhat what we observe in the economy is the result ot a completely different mechanism, and lhe lact that the IS-LM model tits well is a coincidence. But this seems unlikely. The IS-LM model looks like a solid basis on which lo build when looking at movements in activity in the short run. later, we will extend thc model to look at the role ol expectations (Chapters 14 to !7> and the implications ol openness in both lhe goods markets and the financial markets (Chapters 18 to 21). Bui we must lirst understand what determines output in the medium run. This is thc topic of the next four chapters.
SUMMARY
• Thc IS LM model characterises the implications ol equilibrium in both the goods markets and the financial markets.
The federal funds rate in the United States is the corollary of the t overnight cash rate in Australia. See Section 4.4 in Chapter 4.
There is no such thing in ► econometrics as learning the exact value of a coefficient or the exact effect of one variable on another. Rather, the best econometrics can do is to provide us with a best estimate—here, the solid line—and a measure of confidence we can have in the estimate—here, the confidence band. For an econometrics refresher, check Appendix 3 at the back of the book.
This is why monetary ► policy couldn't prevent the 2001 slowdown. When, at the start of 2001, the Fed realised that the US economy was slowing down and started cutting the federal funds rate, it was already too late for these cuts to have much effect on output in 2001.
• The IS relation and the IS curve show the combinations ol the interest rate and the level ol output thai are consistent with equilibrium in the goods market. An increase in the interest rate leads to a decline in output. The IS curve is downward sloping.
GOODS AND riNANCIAI MARKETS THE IS IM MODEL
chapter S
• Thc LM relation and thc LA 1 curve show the combinations ol the interest rate and the level ol output consistent with equilibrium in financial markets. C.iven the real money supply, an increase in output requires an increase in the interest rale. The LM curve is upward sloping.
• A liscal expansion shilis the /5 curve to ihe right leading to an increase in output A liscal contraction shilts the IS curve to the Icll, leading to a decrease in output and the interest rale
• A monetary expansion shifts the LAI curve down leading to an increase in output and a decrease in the interest rate. A monetary contraction shilts the LAI curve up. leading to a decrease in output and an increase in the interest rale.
• The combination ol monetary and liscal policies is known as the monetary liscal policy mix, or simply the policy mix.
• Monetary policy can be conducted by fixing either the money stock or the interest rate with most central banks in developed countries opting now lor the latter. The effect ol liscal policy on output is greater when the interest rale is kept fixed.
• Sometimes monetary and liscal policies are used lor a common goal. Sometimes they are not and the monetary-fiscal mix rellects tensions or even disagreements between the government which is in charge ol liscal policy and the central bank which is in charge ol monetary polio
• The IS-LM model appears to describe well the behaviour ol thc economy in the short run. In particular, thc effects ol monetary policy appear lo be similar lo those implied by the IS—I M model once dynamics are introduced in lhe model. An increase in the interest rate due to a monetary contraction leads ю a steady decrease in output, with the maximum ellect taking place alter about eight quarters.
KEYTERMS


monetary contraction monetary lightening, 105
monetary tiscal policy mix policy mix 107 conl idence band, IIS