The transmission mechanism of monetary policy

62 MONETARY BULLETIN 2001/4

1. Introduction

In implementing its monetary policy, the CentralBank of Iceland needs to assess the impact that itsdecisions have and the time considered before theyaffect the economy. This is particularly true now thatthe Bank has adopted a formal inflation target, whichcalls for a forward-looking monetary policy (cf. thediscussion in Pétursson, 2000).

This paper discusses the transmission mechanismof monetary policy, i.e. how changes in the centralbank policy rate are transmitted through the econo-my, affecting aggregate demand, inflation expecta-tions and the rate of inflation. As will emerge, this is

a fairly complex process and the impact may varyfrom one time to another. However, the typicaleffects of monetary policy actions in Iceland and thetime lags before they have any effect appear broadlyspeaking to be consistent with the experience ofother countries.

1.1. Overview of the transmission mechanismThe reason that central banks can affect interest rateformation in the economy is that they have monop-oly power in supplying money in the economy, i.e.base money which comprises notes and coin in cir-culation and the reserves of financial institutions inthe central banks. By setting the price of base money,i.e. the interest rate at which financial institutions canborrow short-term capital from central banks, theycan influence the volume and price of liquidity in thefinancial system, which in turn affects interest rateformation in the economy.

Like its counterparts in other countries, theCentral Bank of Iceland attempts to influence inter-est rate formation in the financial system by adjust-

THÓRARINN G. PÉTURSSON1

The transmission mechanism of monetary policy

The Central Bank of Iceland bases its monetary policy on setting its interest rate in transactions withother financial institutions in the money market, in order to affect the behaviour of individuals andfirms, thereby keeping aggregate demand in line with its growth potential and maintaining inflationexpectations which are consistent with the Bank’s 2½% inflation target. The process through whichmonetary policy decisions affect aggregate demand in the economy, inflation expectations and the infla-tion rate is generally known as the monetary policy transmission mechanism. This paper discusses thisprocess and the lags from monetary policy decisions to its effect on the economy. The findings suggestthat Central Bank of Iceland monetary policy changes are in general first transmitted to domesticdemand after roughly half a year, with a peak effect after one year. Policy first affects inflation after ayear, with a peak effect about 1½ years after the interest rate rise. In the long run, however, monetarypolicy has no effects on the real economy. Broadly speaking this is consistent with other countries’expe-rience. It should be underlined that the transmission mechanism is a complex process which may alterfrom one time to another.

1. The author is division chief of economic research at the EconomicsDepartment of the Central Bank of Iceland, and an assistant professorat Reykjavík University. He would like to thank IngimundurFridriksson, Jón Steinsson, Lúdvík Elíasson, Már Gudmundsson andseminar participants at the Central Bank of Iceland on October 15, 2001for their constructive remarks. The views presented here are those ofthe author and do not necessarily reflect the views of the Central Bankof Iceland.

MONETARY BULLETIN 2001/4 63

ing the interest rate on its repurchase agreementswith financial institutions, which are transactionswhereby the Bank and other financial institutionsswap securities and liquid assets on condition thatthese transactions are reversed after fourteen days.2

Central bank interest rate decisions affect short-and long-term interest rates, liquidity in the financialsystem, the quantity of money and bank credit,exchange rates, other asset prices and, last but notleast, market expectations about the future develop-ment of all these variables. All this, in turn, influ-ences consumption and investment decisions of indi-viduals and firms, and thereby aggregate demandand, ultimately, inflation, as shown in Figure 1.3

It should be emphasised that the following dis-cussion is primarily confined to the economy’s typi-cal reactions to changes in the central bank policyrate. In each case, the ultimate effect of monetarypolicy is determined by whether the measures areforeseen or not, and how they affect expectationsabout future monetary policy actions. The economy

may therefore show different responses from onetime to another, depending upon conditions within itin each case.

2. The transmission of monetary policy throughthe financial system

The first stage of the transmission mechanism is thefinancial system channel. There, monetary policyactions are primarily manifested by changes in short-and long-term interest rates, asset prices, liquidityand the exchange rate of domestic currency.

2.1. Short- and long-term interest ratesA change in the central bank policy rate has animmediate impact in the money market (the marketfor securities with maturity from one day to oneyear). A rise in the policy rate generally leads to animmediate rise in interbank and treasury bill rates,although not necessarily by the same amount; thisdepends among other things on the securities’ matu-rity compared with that of the instrument priced bythe policy rate, and how actively the instrument inquestion is traded. Interest rates on commercialbanks’ short-term instruments should rise relativelyquickly, since they are normally financed to a large

2. For a more detailed description of the Central Bank of Iceland’s instru-ments and implementation of monetary policy, see Kristinsson (2000).

3. Different transmission channels may work in tandem and affect eachother. For simplicity, these interactions are not shown in the figure.

Market interestrates

Inflation

Net exports

Domesticdemand

Central Bankpolicy rate

Exchange rate

Expectationsand credibility

Money andcredit

Asset prices

Aggregatedemand Output gap

Import prices

Figure 1

The transmission mechanism of monetary policy


extent in the money markets. Interest rates on short-term variable-rate loans rise soon afterwards. Rateson short-term fixed-rate instruments also rise, but ingeneral after some lag. Monetary policy generallydoes not have much effect on the spread betweenlending and borrowing rates, so deposit rates shouldalso rise fairly soon after a rise in the policy rate (see,for example, MPC, 1999).

The impact of a policy rate change on long-termnominal rates is not as obvious as on short-term rates.A higher policy rate may cause either a rise or a fallin long-term interest rates, since these are broadlyspeaking determined by the average of current andexpected future short-term interest rates for the matu-rity of the longer instruments (see, for example,Pétursson, 1998a). The impact on long-term interestrates therefore depends on the effect that the policyrate rise has on market expectations about futuredevelopments of short-term rates, and especiallyabout future inflation developments, which are animportant determinant of nominal interest rates. Forexample, if market participants expect short-termrates to fall substantially in the future, long-terminterest rates could even fall in response to a policyrate rise. Such expectations might reflect, for exam-ple, confidence that the interest rate rise is sufficientto prompt a sizeable reduction in the inflation rate inthe future. On the other hand, expectations that thepolicy rate rise is the first of many on the part of thecentral bank may push up long-term interest rates bymore than the short-term rates.

The most common effect, however, is that short-term interest rates will gradually revert to their orig-inal level. This implies that long-term interest rateswill rise, but not by the same amount as short-termrates.4

2.2. Asset pricesA central bank policy rate change also affects assetprices, e.g. equity prices and housing prices. In gen-eral, a rise in the policy rate leads to a fall in equityprices, since the income stream which the share price

measures is now discounted at a higher rate of long-term interest than before. Thus equity will be worthless, all things being equal (especially inflationexpectations). Equity prices can also fall because ris-ing interest rates and decreasing supply of moneyreduce the demand for equity, for example in favourof bonds which now pay higher interest and aretherefore a more attractive investment option (see,for example, Meltzer, 1995).

However, a rise in interest rates need not neces-sarily lead to a fall in share prices. Market partici-pants might interpret the higher interest rate as indi-cating that the central bank foresees faster economicgrowth in the future than had previously beenassumed. In such a case, a rise in interest rates couldpush up equity prices, since expectations of moreeconomic growth and consequently higher corporateprofits would outweigh the direct impact through thediscount factor. The effects of monetary policy onexpectations will be discussed in more detail below.

Generally speaking, a rise in the policy interestrate should also cause a fall in housing prices, sincehousing finance costs will increase with a correspon-ding reduction in demand. All things being equal,less demand for housing ought to lead to smallerrises, or even reductions, in housing prices. The sameapplies to other assets such as land.

2.3. Liquidity, money and bank creditMonetary policy also affects the volume of liquidfunds in the economy (i.e. financial assets that thatcan easily be converted to cash at short notice, bear-ing little or no interest). When the central bank raises its policy rate, the opportunity cost of holdingsuch liquid assets increases, since other interest-bear-ing financial assets have become more attractivecompared with liquid funds. The demand for liquidfunds should therefore decrease.

A higher policy rate can also affect the demandfor broad money. However, this effect is not as obvi-ous, since broad money generally pays interest, atleast in Iceland. Generally speaking, a rise in the pol-icy rate will push up interest rates on both securitiesand broad money. The effect on the opportunity costof money (i.e. the interest rate spread) is thereforeunclear. Nonetheless, all things being equal a rise inthe policy rate will lead to a lower price level andreduce income and wealth (or slow the growth of

4. In Iceland, the relation may be even more complex, since a large partof long-term liabilities have indexed-linked interest rates. If prices aresluggish and there is some substitutability between indexed and non-indexed bonds, a rise in the policy rate should lead to a temporary risein indexed interest rates. See Pétursson (2001b) and the findings in Box1 below.


these aggregates), hence reducing the demand formoney.5

If the demand for money contracts in the wake ofa rise in the policy rate, a corresponding reductiontakes place in bank deposits, which in turn couldimpair the banking system’s lending capacity ifbanks encounter difficulties in funding their lendingin other ways. The cost of financing credit alsoincreases, since banks need to seek other, moreexpensive sources, and lending rates rise as a result.Falling supply of credit and rising lending rates canhave widespread effects on individuals and firms thatdo not have easy access to alternative sources offinancing. This effect is discussed in more detailbelow.

2.4. The exchange rateCentral bank interest rate decisions also have animpact on the exchange rate of the domestic curren-cy. The precise effect is uncertain, however, since itis determined by expectations about subsequentdomestic and international economic developments,including domestic and foreign interest rate andinflation developments. Given market expectations,an unexpected rise in domestic interest rates propor-tionate to comparable rates abroad should in generalcause the exchange rate to appreciate.6 The reason isthat higher interest rates make domestic financialassets more attractive than comparable foreignassets, all other things being equal. Demand fordomestic currency therefore increases, causing a risein its price, i.e. an appreciation.

This appreciation should be large enough for theexpected return on domestic and foreign assets to beequal. Otherwise, unexploited arbitrage opportuni-ties would exist. At any given time, the interest ratedifferential between comparable domestic and for-eign securities should therefore correspond to theexpected change in the domestic currency over thematurity of the investment, plus a risk premium thatinvestors demand for investing in domestic assets

(which for Iceland could reflect, for example, thelong history of depreciations of the króna, the lack ofCentral Bank credibility, the small size of the marketand lesser liquidity of domestic securities relative tocomparable foreign ones).

However, a rise in nominal interest rates whichreflects higher inflation expectations generally causes the domestic exchange rate to depreciate sinceinvestors expect higher future inflation to reduce itsvalue, i.e. cause a depreciation. Therefore they im-mediately sell the domestic currency to avoid ex-change rate losses later. This increased supply of cur-rency then causes it to depreciate. A rise in the poli-cy rate may therefore weaken the domestic exchangerate if it is insufficient to offset higher inflationexpectations, meaning that the central bank’s realpolicy rate has in fact fallen, despite the nominal rise.

2.5. Expectations and confidenceAs mentioned earlier, expectations about the futuredevelopments of, for example, economic growth andinflation are a major determinant of the ultimateimpact of monetary policy actions. Monetary policycan affect these expectations and the confidence withwhich they are held. Such changes in expectationsinfluence the behaviour of financial market partici-pants and other agents in the economy, includingindividuals’ expectations about employmentprospects and firms’ expectations about future salesand profits. Monetary policy can thus affect thebehaviour of individuals and firms through theirexpectations; such effects may even become evidentbefore those which are channelled through the priceand volume of the various financial assets discussedabove. Even expected monetary policy decisions canaffect behaviour, without actually being implement-ed.

However, the precise impact of monetary policyon expectations is difficult to ascertain precisely, andprobably varies from one time to another. In generala rise in the policy rate would be interpreted as sig-nalling a need to slow down the economy in order toachieve the inflation target, with future growthprospects deteriorating and inflation expectationsfalling if the policy action is credible. Such aresponse would reinforce the bank’s efforts to raiseinterest rates to curb excess demand in the economy.If the policy action lacks credbility, e.g. if the public

5. An estimation of money demand in Iceland, finding a negative relationbetween money demand and the opportunity cost of money, is found inPétursson (2001a). See also Gudmundsson (1986).

6. Exchange rate developments are generally characterised by substantialshort-term fluctuations and the factors causing this volatility could eas-ily swamp the short-run effects of monetary policy on the exchangerate.


regards the bank’s efforts to slow down the economyas inadequate, expectations of further interest raterises could arise, which would magnify the contrac-tion impact. A higher policy rate could even be inter-preted as reflecting that the bank views that the econ-omy is growing faster than had earlier been thought,

thereby spurring future growth expectations. Thisresponse could diminish the tightening imposed bythe interest rate rise.

Uncertainty about monetary policy’s effects onexpectations and confidence in the economy therebyincreases the uncertainty surrounding its effects on

The first effects of monetary policy actions are onfinancial markets, particularly those where short-termfinancial assets are traded, e.g. the domestic moneymarket.

As with the monetary policy transmission mecha-nism in general, the time it takes Central Bank of Ice-land actions to have an impact within the financial sys-tem is subject to uncertainties. The lags may dependon factors such as the extent to which market partici-pants foresaw the actions, how they interpret theirimpact on future economic prospects and their predic-tions of the Central Bank’s future actions. Thus thetime lags probably vary from one period to another.

To give some idea of how policy actions workthrough the financial system, the figure below showsthe impact of an unexpected 1% increase in the policyrate on money market interest rates (3-month treasurybills), indexed treasury bond interest rates (5-yeartreasury bonds) and bank lending rates (average inter-est rate on indexed lending). The estimation is based

on a VAR analysis, see Pétursson, 2001b). It repre-sents an estimation of a typical response by these mar-ket rates to an unexpected rise in the policy rate overthe past ten years and should not be interpreted as aforecast of responses to the Bank’s actions in thefuture.

According to this analysis, money market ratesimmediately rise by 0.7% following a 1% rise in theCentral Bank rate. The effect peaks during the firstmonths after the rise, then gradually fades out. Bondrates also rise immediately by roughly 0.2%, animpact transmitted through money market rates. Theeffect peaks during the first months after the rise, thengradually fades out. The analysis shows that indexedlending rates do not rise immediately, but have risenby roughly 0.3% when the effect peaks around fourmonths after the policy rate rise, largely transmittedthrough the bond market. The effect gradually wanesafter that and fades out after around a year, based onthe statistical confidence bands.

Box 1 The effects of Central Bank monetary policy on the financial system

Impact on the financial system of an unexpected 1% increase in Central Bank interest rate

Percentage deviation from baseline with 90% confidence bands

Months Months Months

1 2 3 4 5 6 7 8 9 10 11 12

0.00.20.40.60.81.01.2

-0.2-0.4-0.6

%

1 2 3 4 5 6 7 8 9 10 11 12

0.0

0.2

0.4

0.6

0.8

1.0

-0.2

%

1 2 3 4 5 6 7 8 9 10 11 12

0.0

0.2

0.4

0.6

-0.2

%

Indexed bond market rateMoney market rate Indexed bank loan rate

other economic variables, which underlines theimportance of a credible and transparent monetarypolicy. This point will be discussed in more detailbelow.

2.6. Time lags from policy decisions to the financialsystemThrough their interest rate decisions, central banksdirectly influence other short-term interest rates and,through them, long-term rates, asset prices, the quan-tity of money and the exchange rate. In general anunexpected policy rate hike causes a rise in long-term interest rates, a fall in asset prices, a reductionin the quantity of money and a strengthening of thedomestic currency. These effects are generally trans-mitted fairly quickly, usually within a few months, asshown in Box 1 which discusses the impact of mon-etary policy on the money, bonds and bank lendingmarkets in Iceland. However, the effects vary fromone time to another and are determined by factorssuch as the business cycle, expectations and centralbank credibility.

3. From the financial system to spending deci-sions

The second stage of the transmission mechanismdescribes the impact of monetary policy from thefinancial system and onto the rest of the economy,i.e. how central bank interest rate decisions affectspending decisions by individuals and firms. In turn,these decisions affect aggregate demand and ulti-mately the rate of inflation. Before describing thisprocess, however, the impact of monetary policy onreal interest rates and the real exchange rate need tobe discussed.

3.1. Real interest rates and the real exchange rateAs discussed in the preceding section, a rise in cen-tral bank policy rate generally produces a rise inother nominal interest rates and the nominalexchange rate of the domestic currency. However, itis not these variables that are most important forexpenditure decisions of individuals and firms, butrather the development of real interest rates and thereal exchange rate.7 It is real interest rates, i.e. nom-inal interest rates adjusted for expected inflationthrough the maturity of the investment, rather than

nominal rates, that determine the profitability ofinvestments and other spending decisions.8 Like-wise, it is the real exchange rate, i.e. the nominalexchange rate adjusted for domestic prices relative toforeign prices, and not the nominal exchange rate,that determines the competitive position of domesticfirms.

Extensive research shows that prices and infla-tion adjust slowly. It can therefore be assumed (asalso confirmed by research) that inflation expecta-tions are also sticky (see, for example, the discussionin Taylor, 1995). Thus a rise in nominal interest ratesand the nominal exchange rate will result in a higherreal exchange rate and real interest rates, all otherthings being equal.9 An increase in these variablesdue to a rise in the central bank policy rate, however,will only prove temporary, while inflation and infla-tion expectations are adjusting to the new interestrate and exchange rate levels. As discussed below,this means that the effect of monetary policy on realvariables such as growth, employment and the cur-rent account can only be temporary.

3.2. Spending decisions of individualsBy influencing interest rates, asset prices, theexchange rate of the domestic currency and the quan-tity of money, monetary policy has an effect on indi-viduals’ behaviour through various channels. One ofthe most important effects of monetary policy isprobably through disposable income. By influencingmarket interest rates, monetary policy affects theinterest rate on savings, as well as on outstandingshort-term liabilities (for example borrowing oncredit cards and overdrafts) and long-term liabilitieswith variable interest rates, which are a relativelycommon debt instrument in Iceland. A rise in the pol-icy rate thus reduces disposable income of netdebtors, which in turn affects their spending deci-sions.10 Higher interest rates prevent individualsfrom maintaining the same level of spending on con-


7. Nominal interest rates and the nominal exchange rate can be importantfor spending decisions of individuals and firms if their access to finan-cial markets is restricted in some way, since these nominal aggregatesaffect their cash flow.

8. Thus it is the real policy rate that determines the tightness of the mon-etary stance, and not the nominal policy rate.

9. As seen in the chart in Box 1, where interest rates on bonds and banklending are real interest rates.

sumption without incurring more debt or drawing onsavings, which in both cases has become moreexpensive. All things being equal, individuals’ con-sumption expenditure ought to contract when interestrates rise. The rise in interest rates does not affectlong-term debts with fixed interest until they reachmaturity, but all new debtors are immediately affect-ed by the higher interest rates.11

Monetary policy also affects the timing of con-sumption decisions, since interest rates in effect rep-resent the price of current consumption relative tothat in the future. When interest rates rise, currentconsumption becomes more expensive compared tofuture consumption, i.e. current saving. Individualsshould therefore reduce their current consumption bya corresponding amount.

Thirdly, monetary policy affects individuals’wealth. As discussed above, a rise in interest ratesgenerally leads to a fall in stock and housing prices.Since these assets constitute an important part ofindividuals’ aggregate wealth, their consumptionexpenditure should decrease, since they are no longeras wealthy. Likewise, their access to credit becomesmore difficult, because housing is often used as col-lateral for loans. Since the market prices of theirassets has fallen, their borrowing capability de-creases. The impact of monetary policy on the accessto credit for individuals and firms is discussed inmore detail below.

Monetary policy can also affect the consumptionexpenditure of individuals through consumers’expectations about their future income and employ-ment prospects. For example, if individuals expectthat a tighter monetary policy stance will reducemedium-term economic growth, they are likely to cutback their expenditure on consumption and their cur-rent indebtedness in order to sustain future consump-

tion more easily. However, a tighter monetary policycould be interpreted as signalling that the rate of eco-nomic growth is faster than had been thought.Individuals’ expectations could thereby be kindled,and likewise their willingness to spend. This effect istherefore uncertain and probably varies from oneperiod to another.

Monetary policy also affects individuals’ con-sumption patterns through the exchange rate channel.An appreciation makes imported goods and servicesrelatively cheaper. The effect is to reduce relativedemand for domestic goods and channel it out of theeconomy, temporarily weakening the competitiveposition of domestic businesses, as will be discussedlater. An appreciation may also affect total consump-tion, although its main impact is probably on thecomposition of consumption expenditure, if forexample a major part of individuals’ assets or liabili-ties is denominated in foreign currencies. Thus a cur-rency appreciation would reduce the indebtedness ofindividuals who have part of their liabilities in for-eign currencies, since their income is generally indomestic currency. Total wealth of these individualswould therefore increase, and with it consumptionexpenditure through the wealth effect. This may soft-en the impact of monetary policy if net foreign lia-bilities form a major part of total individual debt. Onthe other hand, the positive wealth effect of an appre-ciation which is prompted by a rise in interest rates isprobably offset by the fact that it entails a greaterlikelihood of a subsequent depreciation, therebyincreasing the currency risk faced by individualswith foreign liabilities.

Finally, monetary policy affects consumptionexpenditure of individuals through their access tocredit (especially for financing expenditure on con-sumer durables such as housing and motor vehicles).The possibility that a decrease in money supply fol-lowing a rise in central bank policy rates couldreduce bank deposits and thereby the lending capac-ity of banks has already been pointed out. In general,individuals have few other finance options than bor-rowing from banks. Other things being equal, indi-viduals’ scope for financing their consumption alsodiminishes. However, it seems likely that the impor-tance of this bank lending channel has decreasedwith the more diverse finance options available tobanks in recent years.


10. Net indebtedness of Icelandic households was estimated at more than160% of disposable income in 2000. Based on a rough evaluation ofinterest-bearing financial assets (excluding equity holdings and pensionfund assets), Icelandic household debt net of interest-bearing assetscould be in the region of 60-70% of disposable income.

11. Disposable income of individuals who are net savers obviously in-creases when interest rates rise. Other things being equal, their propen-sity to spend ought to increase. Nonetheless, studies show that the over-all impact of higher interest rates is generally a contraction in consump-tion expenditure, although this is probably only small at first. As dis-cussed later, the second-round effects, i.e. when aggregate demand iscontracting, are more important (see, for example, MPC, 1999).

Monetary policy has, however, another and moreimportant effect on individuals’ access to creditthrough the banking system, called the balance-sheetchannel. With a rise in the policy rate, the net worthof individuals diminishes since their balance sheetshave deteriorated (e.g. through the above-mentionedrise in interest expense on outstanding current liabil-ities and the negative wealth effect). Such circum-stances exacerbate the problems of adverse selectionand moral hazard.12 By reducing the banking sys-tem’s willingness to lend, this reduces individuals’spending ability (see, for example, Bernanke andGertler, 1995). In general lending ability within thebanking system also diminishes, since the interestrate rise has the same effect on its net worth andaccess to new credit. Monetary policy can also affectindividuals’ willingness to borrow through a liquidi-ty channel, because a rise in interest rates canincrease the probability of financial distress later on.They therefore reduce their demand for assets whichcould be difficult to liquidate at short notice, such ashousing and consumer durables, but increase theirdemand for highly liquid assets such as bank depositsand securities (see, for example, Mishkin, 1978).

The impact of monetary policy is therefore trans-mitted to individuals’ spending decisions through awide range of channels. In general a rise in the cen-tral bank policy rate will cause a contraction in totalexpenditure by individuals. Furthermore, they willtend to switch their spending from domestic to for-eign goods in response to the exchange rate effectwhich makes imports relatively cheaper. The extentof the effect of monetary policy on individuals’expenditure decisions, and even sometimes on thedirection they take, may vary from one time to anoth-er, depending upon factors such as its effect on indi-viduals’ expectations and confidence.

3.3. Spending decisions of firmsMonetary policy also affects firms’ spending deci-sions through interest rates, asset prices, theexchange rate and the quantity of money. This effectmay vary according to the nature of the business, thesize of the firm and its sources of finance.

Firms relying on bank financing, or other types ofcredit funding linked to domestic short-term interestrate developments, are affected directly. Increasedborrowing costs reduce their profit and raise therequired return on all new investments. The willing-ness and ability to embark on new projects is there-fore diminished, other things being equal. Likewise,higher interest rates increase firms’ inventory costs,which are often financed with short-term credit.Higher interest rates also affect the demand for la-bour, by reducing their willingness to hire new staff.Firms might even reduce employment or hoursworked.13

Monetary policy also affects the cost of capitalthrough its temporary impact on long-term real inter-est rates. Thus a rise in the policy rate should alsoprompt a temporary rise in the required return onnew projects, making it more likely for firms to post-pone or simply abandon such plans.

However, it should be borne in mind that somefirms are probably little affected by changes in thepolicy rate. For example, those with minimal liquidassets or short-term liabilities will have their cashflows largely unaffected. The same applies to firmswhose liquid assets and short-term liabilities areroughly matched. Nonetheless, they will be affectedby monetary policy changes through long-term realinterest rates whenever they need to use the capitalmarkets to fund long-term investments. This is lesstrue of large firms or those able to raise funds ininternational financial markets, because they are lessdependent on funding through the domestic financialsystem and domestic interest rate developments.They will, however, have greater exchange rate riskif they have foreign-denominated liabilities and partof their revenues or assets are in domestic currency.


12. Adverse selection refers here to an increase in the average risk of lend-ing following a reduction in the net worth of borrowers, whereby low-risk borrowers defer or back out of borrowing due to a rise in the inter-est rate and thereby the required return on the underlying decision.Moral hazard refers here to an increase in the average risk of lending asborrowers are prepared to take more risk when their net worth de-creases, since they have a lower equity stake in the project, giving themincreased incentives to engage in risky investments.

13. However, higher interest rates can have a positive impact on firmswhich have favourable liquidity positions, since they improve theircash flow. These firms might therefore increase their investment fol-lowing a rise in interest rates. But it is also possible that these firms willdecide to channel their extra cash flow into financial assets or, say,higher dividend payments to shareholders.

Another effect of monetary policy on firms’spending decisions is through the asset price channel.As mentioned above, a rise in the policy rate usuallyleads to a fall in equity prices, causing the marketvalue of firms to fall relative to the replacement costof capital (a fall in the q-ratio, see Tobin, 1969).Accordingly, it becomes relatively more expensivefor firms to issue new equity to finance new invest-ments.

Firms experience a similar wealth effect to indi-viduals. Higher interest rates reduce their net worthand their cash flow worsens, as well as lowering theprice of assets which can be used as collateral. Firmscould then encounter problems in funding new proj-ects. Adverse selection and moral hazard problemsincrease accordingly, making the banking systemeven less willing to lend (this effect is known as thefinancial accelerator effect). This applies particularlyto small and new firms which lack easy access toother sources of financing outside the banking sys-tem (see Bernanke and Gertler, 1995).

Exchange rate changes are another importantinfluence on firms’ spending decisions. A temporaryreal appreciation of the domestic currency worsensthe competitive position of domestic firms that havecosts in domestic currency but produce for exportmarkets or in competition with imports. They need tocounteract the appreciation either by reducing theprices of their products, which leads to a lower prof-it margin, or by keeping their prices unchanged,which leads to a lower market share. Various servicessuch as tourism also experience a contraction in thewake of an exchange rate appreciation. Monetarypolicy can also affect firms which produce solely forthe domestic market or firms that are not in interna-tional competition if they have, for example,unhedged foreign debt positions. By raising the networth of firms with net foreign debt, a strongerexchange rate could also increase their willingness tospend. Counteracting this is the greater probability ofa subsequent depreciation, which would increasetheir foreign exchange risk. Fluctuating exchangerates would, in addition, increase the risk of invest-ing in such firms, leading to higher financing coststhan otherwise.

Finally, monetary policy has an effect on firms’expectations and confidence about the economic out-look, just as it does for individuals. This can be a cru-

cial factor, especially for long-term irreversibleinvestments. Expectations about future sales, interestrate developments and future risk play a major role inthe timing of investments. If a firm foresees a con-traction in the near future, optimism about futuresales will be dampened, making the investment lessattractive. Increased uncertainty about the future isalso important, since the greater the uncertainty, theriskier the investment, with a corresponding increasein required rate of return. The impact on expectationsis difficult to predict for firms, just as in the case ofindividuals, and probably varies from one period tothe next.

Monetary policy therefore has an impact on cor-porate expenditure decisions through various chan-nels. In general, a rise in central bank policy rate willcause a contraction in firms’ activities. However, thescale of the impact is difficult to predict, and it cansometimes operate in the opposite direction. It isprobably determined among other things by changesin expectations and confidence.

4. From expenditure to demand and inflation

As shown in Figure 1, the next stage of the transmis-sion mechanism describes how individual and busi-ness expenditure decisions affect aggregate demand,i.e. how aggregate demand is affected by monetarypolicy when expenditure decisions by all individualsand firms are added up to determine total expendi-ture.14 Changes in spending decisions by individualsand firms then affect the rest of the economy, eventhose agents not directly affected by monetary poli-cy.

4.1. Aggregate demandAs discussed in the previous section, a rise in the pol-icy rate causes at least some individuals and firms toreduce their expenditure on consumption and invest-ment. Total expenditure contracts, with a correspon-ding drop in aggregate demand.


14. Aggregate demand is defined as national expenditure and net exports,where national expenditure is the sum of private consumption expendi-ture, government consumption expenditure and investment spending,and net exports are the difference between exports and imports of goodsand services. Aggregate demand is therefore equivalent to the grossdomestic product at market prices (GDP).

When individuals reduce their expenditure, thismeans that their private consumption decreases (orgrows at a slower rate). The same applies to theirinvestments in residential housing and other durablegoods. All these factors cause private consumptionand thereby aggregate demand to decrease (or growat a slower rate). Likewise, individuals shift theirdemand from domestic goods to imports, which arerelatively cheaper after the domestic currency hasappreciated. Demand is thereby channelled out of theeconomy, boosting imports and also slowing downgrowth of aggregate demand.

Firms also reduce their outlays, so that theirinvestments and activities decrease (or grow moreslowly). The same applies to inventories and variousexpenditure decisions such as maintenance projectsand dividend payments to shareholders. All thesefactors combine to reduce the growth rate ofdemand.

Lower aggregate demand has an impact on firmsand individuals that were unaffected directly byhigher interest rates. For example, demand for theproducts of a firm that was not directly affected bythe interest rate rise could contract when privatespending declines. Demand could also shrink if itsproducts are used as inputs by other companieswhich are directly affected by the interest rate rise.Likewise, the disposable income of individuals whowere not affected by higher interest rates coulddecrease if, for example, the firms they work for areforced to cut back operations in response to lessdemand for their products. The fact that these sec-ond-round effects can to some extent be anticipatedalso affects expectations, amplifying the total effect.A manufacturer, for instance, could cut back produc-tion because he expects decreasing demand in thenear future following a tighter monetary policy, evenif he is not directly affected by the interest rate rise.This decision will then affect other firms supplyinginputs that it uses for its production.

Firms and individuals thereby scale down theirown activity because they expect a general contrac-tion throughout the economy, even though theyremain unaffected directly by the higher central bankpolicy rate. Lower demand can then reduce the networth of businesses and individuals, curtailing evenfurther their ability to raise finance. These second-round effects of monetary policy on demand proba-

bly weigh heavier than the direct first-round effectsdiscussed above.

4.2. Demand, inflation and inflation expectationsThe above discussion examined how monetary poli-cy can temporarily affect aggregate demand. AsFigure 1 shows, central banks can ultimately exert animpact on domestic inflation through this demandchannel. More specifically, the demand channeloperates through the output gap, which is the differ-ence between the actual level of production and thepotential output of the economy, i.e. the level of pro-duction where domestic firms are operating undernormal capacity utilisation (see further below).When aggregate demand exceeds potential output, apositive output gap develops whereby firms operateat a capacity level above their normal capacity levels,and output is above its sustainable level. This is, forexample, reflected in excess demand for labour, lead-ing to upward pressures on wage costs. Firms pass onpart of their extra costs to prices, fuelling inflationarypressure. Excess demand for their products alsogives firms an opportunity to raise their markup.Conversely, a negative output gap eases wage andinflationary pressures.

An increasing output gap during economicupswings is generally accompanied by escalatinginflation, as can be seen from in Figure 2 whichshows the development of the output gap and infla-tion in Iceland since 1991.15 There, a negative outputgap during the first half of the decade went hand inhand with falling inflationary pressures. In the pastfew years, however, the output gap has widened andinflationary pressures have built up.

The output gap is thus an important indicator forthe future development of inflation.16 Only at thelevel of demand where actual output corresponds toits potential level can inflation be stable. It is impor-tant to realise that this level of inflation can be eitherhigh or low, depending on inflation expectations.Holding actual output at its potential level will sim-


15. The figure shows the average of different estimation methods of theoutput gap. More detailed information is given in Monetary Bulletin2000/4, pp. 14-15. The forecast is based on the Central Bank’s lastinflation forecast and the National Economic Institute’s forecast fromOctober 1, 2001.

16. This is confirmed in an econometric study for Iceland in Pétursson(2001c).


ply give the level of inflation that agents expect, asreflected in wage settlements and product prices.

Even though maintaining actual output at itspotential level can be compatible with high or lowinflation (as long as it remains steady), the cost ofhigher inflation is much greater, since even whensteady and foreseen it entails considerable sociallosses, including distortive taxation. The role ofmonetary policy is therefore to maintain actual out-put close to its potential level and a low, steady rateof inflation. This aim can only be achieved if mone-tary policy is considered credible, so that inflationexpectations reflect the Central Bank’s inflation tar-get.

4.3. The exchange rate, import prices and inflationFigure 1 shows how monetary policy can also affectdomestic inflation through the direct effect of theexchange rate on imported inflation. This arisesbecause imported goods are an important componentof domestic prices, both via imported consumergoods and services and as a major cost item fordomestic producers who use imported inputs in theirproduction. The exchange rate can therefore directlyaffect prices of these goods by affecting their price indomestic currency. A rise in the policy rate whichstrengthens the domestic currency can cause a directreduction in domestic prices (or the rate at whichthey rise) by reducing the domestic currency price ofimported goods and services. This effect is obvious-ly more important the more open that the economy isfor international trade and the more dependent it ison imported consumer goods and services. A number

of studies have confirmed the historical importanceof this channel for Iceland (e.g. Gudmundsson, 1990;Andersen and Gudmundsson, 1998; and Pétursson,1998b, 2001c).

While this direct exchange rate channel wouldgenerally reduce the time lags from policy decisionsto the final effects on inflation, some time may passbefore the effect on prices of imported goods istransmitted to final prices to consumers. Theexchange rate effect needs to pass through a numberof intermediaries from importers to retailers, each ofwhom may respond differently. Moreover, the effectmay vary from one time to another, depending on thenature of the underlying exchange rate shock. Forexample, if importers regard the exchange ratechange as being temporary, it can be sensible forthem to absorb it through their markup instead ofpassing it on to consumers with the risk of losingmarket share, besides the fact that frequent pricechanges can be costly to implement (see e.g.McCarthy, 1999). Likewise, importers can invest indifferent financial products to hedge against short-term exchange rate fluctuations. An importer wouldthus not pass through an exchange rate change toprices until it was obviously permanent.

The size of the exchange rate pass-through alsodepends on how much competition prevails at differ-ent levels of the import chain. Strong competitionamong intermediaries makes it more difficult for anysingle one to pass on an exchange rate change, sincehe would risk losing market share if the others do notraise their prices as well.

Another crucial factor may be the level of domes-tic demand. Robust domestic demand growth makesit easier for intermediaries to pass on a exchange ratedepreciation to prices without the risk of losing mar-ket share. On the other hand, this is much more diffi-cult if the economy is in a recession where importersare likely to be forced to absorb some of theexchange rate depreciation through their markup.

The exchange rate pass-through probably variesfrom one period to another. Nonetheless, Icelandicand international experience suggests that the pass-through has weakened somewhat in the past fewyears (see e.g. Sveriges Riksbank, 2001). Variouspossible reasons have been cited. Global competitionhas probably decreased firms’ scope for passingthrough exchange rate changes to prices. Similarly,

Figure 2

Inflation and the output gap (%)

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

0

2

4

6

8

-2

-4

-6

%

Forecast

Inflation

Output gap

firms now have better techniques for hedging againstshort-term exchange rate fluctuations. Exchange rateflexibility has also increased in Iceland and manyother countries. The public probably views changesin the exchange rate differently when it can fluctuatefrom one day to the next, compared with a periodwhen changes are rare. Infrequent exchange ratechanges are probably interpreted as being permanent,which will naturally be quickly passed through todomestic prices. The more frequent that exchangerate changes are, the more likely they are to be inter-preted as temporary. Such changes do not need to bereflected in prices until it is clear that they are per-manent, as pointed out earlier.

Thus the impact of short-term exchange rate fluc-tuations on domestic prices probably decreases withgreater exchange rate flexibility and increasingfinancial innovations. Permanent changes in theexchange rate, on the other hand, will ultimately bepassed through to prices. Nonetheless, it is importantto distinguish between the price level impact andinflationary impact of exchange rate changes. A per-manent depreciation will lead to a permanent rise inprices in the long run. While prices are adjustingtowards the new steady state, inflation will emerge.This inflation will only be temporary while theadjustment towards the new steady state takes placeand in the long run the inflationary effect will disap-pear, assuming that the exchange rate depreciationdoes not alter long-term inflation expectations.

4.4. Time lags from policy decisions to the economyAs pointed out above, it can take a considerable timefor monetary policy to affect economic activity.Changes in the policy rate generally affect prices offinancial assets in the domestic money market veryquickly, which then pass through to prices of otherfinancial assets, such as the exchange rate, equityprices and long-term bond prices. However, it cantake considerably longer for these price changes toaffect expenditure decisions by individuals andfirms. For example, some time may elapse beforeinterest rate changes begin to affect the repaymentsburden on long-term loans, such as housing loans,and even longer before their impact is felt on indi-viduals’ spending behaviour. Similarly, changes inindividuals’ spending behaviour can lead to a buildupin retail inventories. The retailer responds by cutting

back on orders from suppliers, who then need toreduce production. This can then result in a lowerlevel of activity and falling demand for labour, whichleads to a reduction in the disposable income of indi-viduals. All these changes take time to emerge.

Empirical evidence from industrialised countriessuggests that, on average, it takes up to six monthsfor a change in monetary policy to affect domesticdemand, with a peak effect after roughly 1-1½ years.These studies also suggest that it takes up to a yearfor a change in monetary policy to affect domesticinflation, with a peak effect after roughly 1½-2 yearsor even later (see e.g. MPC, 1999 and Viñals andVallés, 1999).

Box 2 presents a rough estimation of the mone-tary policy lags for Iceland. It shows that the lags andthe orders of magnitude are more or less the same forIceland as other industrial countries. The impact oneconomic activity is first felt around half a year aftera rise in Central Bank interest rates, with a peakeffect after roughly one year. The initial impact ondomestic inflation appears around one year after theinterest rate rise, with a peak effect after 1½ years.The time lags from policy decisions to inflationmight therefore be somewhat shorter in Iceland thanin most other industrial countries, which couldreflect the relative importance of the exchange ratepass-through.

It should be emphasised that any analysis of thelags in the transmission mechanism of monetary pol-icy is subject to great uncertainty, both in Iceland andabroad. The uncertainty is probably even greater inIceland since a relatively short data period underliesthe econometric estimates, besides the fact that majorreforms have been made to Iceland’s institutional andmonetary policy framework over the past decade. Inaddition, the transmission lags probably vary fromone period to another, governed by factors such asthe business cycle, expectations and the credibility ofthe Central Bank. Monetary policy is thus likely towork through long, variable and uncertain lags whichare difficult to determine precisely. The above esti-mation of the transmission lags therefore only pre-sents the economy’s typical responses to monetarypolicy decisions at “normal” times. As discussed inthe following section, conditions may arise in whichthe economy responds in a completely different way.



The time that it takes for Central Bank of Icelandinterest rate changes to have an impact on economicactivity and inflation is subject to uncertainties in thesame way as the financial market pass-through. Itcould even be argued that this uncertainty increases,the further we travel along the transmission mecha-nism. As the main text suggests, the time lags dependon factors such as the extent to which agents foresawthe actions, how they interpret their impact on futureeconomic prospects and their predictions of theCentral Bank’s future actions. Thus the time lagsprobably vary from one period to another.

The figure above shows an estimation of the typi-cal impact of an unexpected 1% rise in the policy rateon the exchange rate (log of the effective exchangerate index), employment (as a proportion of the labourforce) and prices (log of the CPI). The estimation isbased on a VAR analysis which includes foreignshort-term interest rates (weighted average) and inter-national commodity prices in order to adjust for theimpact of supply shocks, plus the above variables. Itis based on quarterly data covering the period from1989 to 2000. The figure only shows a typicalresponse of the economy to an unexpected rise in thepolicy rate over the past ten years and should not be

interpreted as a forecast of responses to Bank’sactions in the future.

According to this analysis, monetary policy haslittle initial impact on the exchange rate of the króna,employment and prices. After a while the króna grad-ually begins to appreciate (as the exchange rate indexfalls) and the maximum impact has emerged after justunder one year, with an appreciation by roughly 0.2%from the baseline. From then on the króna starts todepreciate again, which is broadly consistent withstandard theories of the relationship betweenexchange rates and interest rates, i.e. a positive inter-

est rate differential vis-à-vis abroad corresponds to anexpected future depreciation of the exchange rate.The initial appreciation of the króna takes place some-what later than in standard models, but is consistentwith the findings of Eichenbaum and Evans (1995)for other countries. The estimation results are, how-ever, subject to considerable uncertainty and theexchange rate effect is not found statistically signifi-cant from zero. This probably reflects the relativelyshort data period on which the estimation results arebased (the data only cover 1-1½ business cycles).Also, for most of the period monetary policy wasbased on an exchange rate peg. A depreciation of the

Box 2 The effects of Central Bank monetary policy on the economy

Impact on the economy of an unexpected 1% rise in Central Bank interest rate

Percentage deviation from baseline with 90% confidence bands

Quarters Quarters Quarters

Employment as a percentageof labour force

Exchange rate Price level

1 2 3 4 5 6 7 8

0.00.20.40.60.81.01.2

-0.2-0.4-0.6-0.8

%

1 2 3 4 5 6 7 8

0.0

0.1

-0.1

-0.2

-0.3

%

1 2 3 4 5 6 7 8

0.00.10.20.3

-0.1-0.2-0.3-0.4-0.5-0.6

%

4.5. Non-linear and asymmetric effectsSo far this discussion has been confined to the econ-omy’s typical responses to changes in monetary pol-icy. Under certain circumstances, however, theseresponses can be fundamentally different.

Some studies imply, for example, that interestrates rises and cuts can have different real effects.Downward stickiness of nominal prices and wages,for example, could cause interest rate rises to havegreater real effects than equivalent interest rate cuts,i.e. an interest rate rise would have proportionallygreater contractionary effects than an equivalentinterest rate cut would have an expansionary effect,given the economy’s position in the business cycle(see e.g. Cover, 1992).

Monetary policy can also have an asymmetriceffect on expectations. For example, higher interestrates could cause the domestic currency either tostrengthen or weaken, depending on the effect onexpectations. There are two effects working againsteach other. All things being equal, a rise in interestrates should slow down the economy and therebyreduce the probability of future inflation. This shouldmake domestic financial assets more attractive andthereby strengthen its exchange rate. Offsetting thisis the possibility that by slowing down the economy,higher interest rates could weaken the exchange rateby reducing the number of attractive investmentopportunities. The final effect of higher interest rateson the exchange rate depends on which of these twoeffects proves stronger. Empirical studies indicatethat the former effect is generally stronger and thatcurrencies rarely depreciate following a monetarypolicy tightening (see e.g. Zettelmeyer, 2000).

The effects of monetary policy on the real econo-my can also depend on its position in the business

cycle. For example, research suggests that the realeffects of monetary policy may be proportionallygreater during recessions than equivalent actions dur-ing booms (see e.g. Garcia and Shaller, 1995). Themain reason is the reliance of individuals and smallfirms on easy access to credit, as discussed earlier.Since credit is generally tighter during recessions,monetary policy can both help to steer the economyout of the recession with interest rate cuts which cre-ate easier access to credit, and also push the econo-my into further recession if the monetary stance hap-pens to be tightened during a recession. The reason isthat the increased tightening would squeeze access tocredit for firms and individuals that rely on fundingthrough the banking system, which could cause prob-lems for them and even increase the probability ofbankruptcies. Both these effects are probablystronger than the effects of a corresponding tighten-ing during a boom, when access to credit is mucheasier.17

The impact on the real economy may also dependon the size of the interest rate change. Menu costmodels, for example, indicate that small changes indemand have little impact on firms’ price decisions,since it can be expensive to adjust prices. Largedemand shocks, however, may lead firms to decide todo so. In this case, minor changes in the central bankpolicy rate would have strong real effects, but large


króna was countered by interest rate rises whichwould suggest a negative relation between interestrates and the exchange rate if the causation was mis-interpreted.

The interest rate rise gradually leads to loweremployment, with a statistically significant effectoccurring after roughly half a year. The effect peaksafter one year when unemployment has increased by0.15% of the labour force (deviation from baseline).

The effect gradually decreases and has vanished afterroughly 1½ years, based on the confidence bands. Astatistically significant effect on prices occurs roughlyone year after the interest rate rise. The effects peaksafter about 1½ years when prices have fallen byroughly 0.3% from the baseline. This corresponds to amaximum impact on the annual rate of inflation aftersome 15 months, when inflation has fallen by 0.35percentage points from the baseline.

17. The effect of tighter money during a recession could be even greaterstill if accompanied by disinflation which raises the real indebtednessof net debtors. All things being equal their net worth deteriorates andbankruptcies become more likely, amplifying the recession. Bernanke(1983), for example, argues that Federal Reserve actions at the onset ofthe Great Depression, and the disinflation in its wake, played a key rolein turning a potentially relatively small contraction into the deep crisisthat actually occurred. There is much less probability of such a scenarioduring a boom.

interest rate changes proportionally smaller effects(see e.g. the empirical study by Ravn and Sola,1996). However, the effects might conceivably be thecomplete opposite. A large interest rate rise could bemore likely than a small one to turn the economy intoa recession, increasing the likelihood of bankruptciesamong debtors, and possibly even breaking thefinancial system. Likewise, a large interest ratechange can affect the credibility of monetary policy,which could amplify the real effects of monetary pol-icy. In both cases the real effects of a large interestrate change would be proportionally greater than thatof a smaller one. It is therefore unclear whether largeor small changes in monetary policy have relativelygreater real effects, but the effects are not necessari-ly symmetric.

4.6. The long-run effects of monetary policyMonetary policy affects aggregate demand in theshort to medium term. In the long run, however, thetrend level of output is determined by its potentiallevel, i.e. the output level that is compatible with nor-mal capacity utilisation. This level defines the long-term growth path of output, where firms have noincentive to change their production decisions andproduct prices change at the rate of expected infla-tion.

Potential output is determined on the supply sideof the economy, for example by available technolo-gy, production factors, the size and skill of the labourforce, the flexibility of the market system and theinstitutional set-up in the economy. The governmentcan influence this level of production, for examplewith changes in the tax structure, by changing theregulatory framework of the economy, and byimproving the functioning of the market system.Monetary policy cannot, however, influence thelong-run capacity level of the economy. If a centralbank tries to maintain output above its potential,aggregate demand will eventually outstrip potentialoutput and inflationary pressures emerge. This couldeven end up in hyperinflation with the severe socialcosts that usually accompany such episodes.

Monetary policy can thus affect the real economyin the short and medium term. Bad monetary policycan even be harmful to the real economy in the longterm, by reducing the effectiveness of the marketmechanism and creating increasing uncertainty. The

general rule, however, is that with a relatively lowand stable rate of inflation, monetary policy can onlyhave long-term effects on nominal aggregates suchas inflation, nominal interest rates and the nominalexchange rate. It cannot have permanent effects onthe long-term growth of real variables.18 Rather, inthe long run, monetary policy determines the mone-tary value of these variables, i.e. the general pricelevel. Inflation therefore indicates how their mone-tary values change over time, namely how the pur-chasing power of money changes over time. It is inthis sense that inflation is a monetary phenomenon(See e.g. MPC, 1999 and Viñals and Vallés, 1999).

5. Conclusions

This paper discusses the transmission mechanism ofmonetary policy, i.e. how changes in the policy rateare transmitted through the economy and affectaggregate demand, inflation expectations and infla-tion.

Empirical evidence from industrial countries sug-gests that it takes up to six months for a change inmonetary policy to affect domestic demand, with apeak effect after roughly one year. It takes up to ayear for a change in monetary policy to affect domes-tic inflation, with a peak effect after roughly 1½-2years. In the long run, however, monetary policyonly affects nominal variables and cannot maintainoutput growth above the growth rate of potential out-put. Attempts to do so will eventually only lead topersistent and even accelerating inflation. The lagsand orders of magnitude appear roughly the same forIceland. However, the effects on inflation appear totake a somewhat shorter time, which might be due tothe relative importance of the exchange rate pass-through via imported goods and services. The rela-tively long lags from the policy rate decisions to theireffects on the economy mean that monetary policy atany time needs to be forward-looking and based oninflation prospects for the coming 1-2 years ratherthan on the current inflationary developments. Withthe new monetary policy framework in Iceland, sucha framework has been formalised.


18. A well formulated monetary policy can, however, reduce the volatilityof real variables, such as the output gap.


The transmission mechanism and its length, how-ever, may change from one time to another, and theeffects can furthermore be non-linear or even dependon whether the central bank is tightening or easingthe monetary stance and on the position of the busi-ness cycle. Considerable uncertainty thus surroundsthe transmission mechanism and the ultimate effects

of monetary policy. To a large extent, they willdepend on how policy affects expectations of eco-nomic agents and the confidence with which theyhold these expectations, further underlining theimportance for monetary policy to be transparent andcredible.

Bibliography

Andersen, P. S., and Gudmundsson, Már (1998), “Inflation anddisinflation in Iceland”, Central Bank of Iceland Working Papers,no. 1.Bernanke, B. S., (1983), “Nonmonetary effects of the financial cri-sis in the propagation of the Great Depression”, AmericanEconomic Review, 73, 257-276.Bernanke, B. S., and M. Gertler (1995), “Inside the black box: Thecredit channel of monetary policy transmission”, Journal ofEconomic Perspectives, 9, 27-48.Cover, J. P., (1992), “Asymmetric effects of positive and negativemoney-supply shocks”, Quarterly Journal of Economics, 107,1261-1282.Eichenbaum, M., and C. L. Evans (1995), “Some empirical evi-dence on the effects of shocks to monetary policy on exchangerates”, Quarterly Journal of Economics, 110, 1975-2010.Garcia, R., and H. Schaller (1995), “Are the effects of monetarypolicy asymmetric?”, University of Montreal, Working PaperSeries, no. 95-6.Gudmundsson, Gudmundur, (1986), “Peningamagn og vextir”,Fjármálatíðindi, 33, 95-104. (Money and interest rates).Gudmundsson, Gudmundur, (1990), “Tölfræðikönnun á verð-bólgu á Íslandi”, Fjármálatíðindi, 37, 43-53. (Statistical analysisof inflation in Iceland).Heikenstein, L., and A. Vredin (1998), “Inflation targeting andSwedish monetary policy – experience and problems”, SverigesRiksbank, Quarterly Review, 4/1998, 5-33.Kristinsson, Yngvi Ö., (2000), “Implementation of monetary pol-icy and Central Bank instruments”, Monetary Bulletin, 2000/4, 40-48.McCarthy, J., (1999), “Pass-through of exchange rates and importprices to domestic inflation in some industrialised economies”,BIS Working Papers, no. 79.Meltzer, A. H., (1995), “Monetary, credit and (other) transmissionprocesses: A monetarist perspective”, Journal of Economic Per-spectives, 9, 49-72.Mishkin, F. S., (1978), “The household balance sheet and theGreat Depression”, Journal of Economic History, 38, 918-937.

MPC [Bank of England Monetary Policy Council] (1999), “Thetransmission mechanism of monetary policy”.Pétursson, Thórarinn G., (1998a), “Explaining the term structure:The expectation hypothesis and time varying term premia”, Uni-versity of Iceland Institute of Economic Studies, IoES WorkingPaper Series, W98:06.Pétursson, Thórarinn G., (1998b), “Price determination and ration-al expectations”, International Journal of Finance and Econom-ics, 3, 157-167.Pétursson, Thórarinn G., (2000), “Exchange rate or inflation tar-geting in monetary policy?”, Monetary Bulletin, 2000/1, 36-45.Pétursson, Thórarinn G., (2001a), “The representative household’sdemand for money in a cointegrated VAR model”, EconometricJournal, 3, 162-176.Pétursson, Thórarinn G., (2001b), “The transmission mechanismof monetary policy: Analysing the financial market pass-through”,Central Bank of Iceland, Working Papers, no. 14.Pétursson, Thórarinn G., (2001c), “Wage and price formation in asmall open economy: Evidence from Iceland”, Central Bank ofIceland, Working Papers, forthcoming.Ravn M. O., and M. Sola (1996), “A reconsideration of the empir-ical evidence on the asymmetric effects of money-supply shocks:Positive vs. negative or big vs. small?”, University of Aarhus,Working Paper Series, no. 1996-4.Taylor, J. B., (1995), “The monetary transmission mechanism: Anempirical framework”, Journal of Economic Perspectives, 9, 11-26.Tobin, J., (1969), “A general equilibrium approach to monetarytheory”, Journal of Money, Credit, and Banking, 1, 15-29.Viñals, J., and J. Vallés (1999), “On the real effects of monetarypolicy: A central banker’s view”, CEPR Discussion Paper Series,no. 2241.Zettelmeyer, J., (2000), “The impact of monetary policy on theexchange rate: Evidence from three small open economies”, IMFWorking Papers, WP/00/141.

Date post:	03-Jan-2017
Category:	Documents
Upload:	ngokhuong
View:	229 times
Download:	0 times

The transmission mechanism of monetary policy

Documents