4.2 Unit shock to the government expenditure
14 The large scale of the model is another reason and thus the choice of parameters is relatively restricted. An alternative specification of the parameters may generate negative steady-state values or multiple equilibria.
15 With the parameter values specified here, the overall employment, (ln), is approximately 0.36 in the steady-state.
10 20 30
To examine the effects of fiscal policy on an economy with money and banking and highlight the role of money and banking in the dynamic responses of macroeconomic variables to expansionary government spending, here we calibrate the model with a unit shock to the government expenditure. It is conducted based on the interest rate rule with the policy parameters being specified as:
10.15,2 0.5
and
3 0.95.Fig. 2 shows that employment in the production sector rises by 0.14% in response to a 1% point increase in the government expenditure, resulting in a 0.09% increase in output, which is slightly lower than 1%. The rising aggregate demand causes the price level and all the interest rates to increase with the expansionary government expenditure as the Keynesian theory suggests. The increased interest rates result in crowding-out effects on output by lowering consumption and the price of capital, indicating the decline in investment. However, the EFP is lowered by 0.12%. As shown in Eq. (19),
10 20 30
dominates. If the investment and consumption in the economy can be financed externally, the lower EFP will raise the level of investment as well as consumption, thereby offsetting the crowding-out effect of government expenditure. As a result, the drops in consumption and investment are relatively moderate, being 0.12% and 0.07%, respectively.
Therefore, a model which neglects banking and money may overstate the crowding-out effect of the fiscal policy. While the public spending crowds out consumption, the demand for deposits is lowered, which should be accommodated by the proportionate drop in the loan. However, the lower demand for the loan is offset partially by the decreases in investment and employment in the banking sector which are crowded out by the expansionary government spending. The smaller magnitude of the loan rate rise reduces the EFP.
4.3 Unit shock to the growth rate of high-powered money
Bernanke & Gertler (1995) argue that the EFP is countercyclical and that it plays the role of financial
accelerator in the transmission mechanism of monetary policy, which is mainly caused by the balance sheet effect. The exclusion of money in the conventional credit channel literature, however, may neglect the impact of the demand for money on the EFP that may in turn attenuate the influences of the financial accelerator, as argued by GM (2007).
The mixed effects of the expansionary monetary policy on the EFP are evident in Fig. 3. With the calibration on the high persistent growth of money
m0.9, the 1% increase in the growth rate of money causes the EFP to drop initially, to rise afterward to a level above the steady state value, and then to decline gradually to the normal level. The initial drop, which is countercyclical, may be viewed as the “financial accelerator” as in Bernanke & Gertler (1995), but the positive levels in the following periods represent the “financial attenuator” as referred to by GM (2007).Except for the movements in the EFP, all other variables are consistent with the results that GM (2007) report. Consumption increases by 0.44% and output rises by 0.48%. The economic boom drives up the inflation as well as
q , whose effects die away gradually after 2 quarters.
t4.4 Unit shock to the effectiveness of collaterals
Before implementing policies under the financial shock, we need to understand how the financial shock impacts the economy. The calibration results are listed in Fig. 4. Most of the results here coincide with the results that GM (2007) report. Consumption, output and employment decrease upon the impact of the shock. The shock causes these three variables
10 20 30
to drop sharply on impact, and then adjust gradually back to the steady state without any
policy assistance. The recession is accompanied by deflation. This is consistent with what is usually observed after the financial crisis strikes. Most of the interest rates decrease as well, except for the uncollateralized rate.16 The increase in
R may, however, mislead the reactions of governments. In a
tTmodel that does not distinguish between interest rates,
R is the only interest rate that guides the
tT consumption and investment decisions of households and firms. In a model where external financing for private spending is likely, however,R and
tLR are the crucial determinants of consumption and
tB saving behavior. While their movements diverge, the interest rate rule of the central bank that viewsT
R as the benchmark interest rate may be misleading. This finding illustrates the importance of money
tand banking for policy making.
The EFP rises upon the shock, consistent with the statement of Bernanke & Gertler (1995) the EFP is countercyclical and characterizes the worsening asymmetric information problem under the crisis. The
16 The rise in EFP coincides with the data shown in Fig. 1: EFP increases significantly while the crisis strikes.
countercyclical EFP will cause the credit to contract, thereby delaying the recovery of the economy.
The value of capital,
q , decreases at the outbreak of the shock, but rebounds rapidly to a level that is
t greater than the steady state in one quarter, because the decline in interest rates raises the discounted sum of the future marginal product of capital. Notwithstanding the initial decline in capital value is close to that in the real world, the speed of recovery seems to be too fast. A possible reason that accounts for the rapid adjustment ofq is the frictionless capital market where there is no price
t stickiness or costs that the capital adjustment may incur.5 Policies under the financial crisis
Based on the above discussions, the policies under the financial crisis will be examined. In this section, we let
tk 1% in all cases and policies are allowed to react to the financial shock. Instead of determining the optimal monetary and fiscal policies, this section will investigate the effects of various policies on the dynamic responses of the economy and whether they can precipitate the economic recovery. In particular, the assessment in this section will center on the sizes and speeds of various policies. The results reveal that large and rapid government expenditure performs better than small and persistent government expenditure. On the other hand, while an expansionary monetary policy will help the economy recover from the crisis, the effects of the interest rate rule crucially depend on the persistence of the financial crisis. The results correspond to Feldstein’s suggestions on the format of fiscal rescue policies, but casts some doubt on the effectiveness of interest rate10 20 30
rules. As a result, this study can also serve as a justification for the large public spending
policies that most governments all over the world are currently implementing while the effects of interest rate rules die away quickly or are even not strong enough to offset the impacts of the crisis.
5.1 Fiscal policy
Two fiscal policies are examined in this section: a stable government expenditure plan, which is small but persistent, and a drastic cure policy, which is large and less persistent. While the first one stands for normal government expenditure, the second one is conducted to find support or objections to the recent policy plans that most economists have proposed for the economy to recover from the recession following the financial distress.
5.1.1 Small and persistent government expenditure
Here we propose a highly persistent government expenditure and allow the public spending to react 100% to the financial shock. Therefore,
D
g and 1
g 0.9. Similar to Section 4.2, the examination of the fiscal policy is conducted under the persistent interest rate rule. The effects of the expansionary fiscal policy are evident by comparing Fig. 4 and Fig. 5. As shown, the implementation of the government spending can successfully help the economy avoid recessions. Output increase by 0.06%instead of -0.05% when there is no public spending.
However, there is no free lunch. The expansionary public spending makes consumption drop, instead of increasing, by 0.19% and the investment decline by 0.1%, compared with the earlier levels before the policy was implemented of -0.07% and 0.013%, respectively. The increase in the aggregate demand leads to a rise in interest rates and the crowding-out effects.
Nevertheless as output and interest rates rise, the EFP is lowered but remains above the normal level and thereby becomes procyclical. This may dampen the crowding-out effect of expansionary fiscal poliy as discussed. Differing from the “financial attenuator” in GM (2007), the procyclical EFP is caused by the expansionary fiscal policy which drives up the output, but is not strong enough to push down the EFP to the level below the steady-state value.
5.1.2 Large and rapid government expenditure
While the more persistent policy can successfully help the economy recover, most economists urge the government to engage in “quick and large” government expenditures in
10 20 30
response to the financial crisis when there is no room for further interest rate decreases, the
economy has run into recessions for more than 4 quarters, consumption and investment have fallen significantly, and there seem to be no signs for the private sector to recover on its own accord.
Therefore, we propose another large and rapid government expenditure by assuming that
D
g and 3g 0.6
, which can generate a similar price level path to that under the previous policy. With the lowpersistence, the 3% increase in government expenditure in response to a 1% financial shock will die out in 10 quarters approximately. Fig. 6 shows that this policy can stimulate the production immediately to 0.2%, but declines quickly with the less persistent spending. The fast adjustment speed also applies to other variables. While the large government expenditure causes drastic drops in consumption and investment initially, they rise back to the normal level in 5 quarters. Similarly, the tax increases greatly at the beginning, but falls down and remains at the steady-state level after 4 quarters. Interest rates, on the other hand, do not vary much with policies.
The quick recovery of the economy, 5 quarters before the government expenditure expires, provides strong support for the large-scale and rapid implementation of policy. As long as the economy is able or
handle the drastic drop in consumption in the early period, without being adversely affected by the current and future inflationary pressure, this enormous aggregate demand stimulus can help the private sector to accumulate enough income and capital to restore the credit market as well as the economy to its normal level before the crisis hit. Compared with the persistent but relatively small fiscal policy, this policy would be more desirable as people wish to avoid the recession sustaining.
5.2 Monetary policy
Since the monetary policy can be easily implemented, it is usually the first step that the government will take in response to the financial crisis. While the interest rate rule does not seem to be effective, control over the base money is also exercised. This finding follows along the same lines as the development of US policies after the outbreak of the financial crisis. While the Fed has consecutively cut the Fed fund rate target from 5.25% to 0.25% within 5 quarters, there do not seem to be any signs of a recovery. In March 2009, the chairman of the Fed, Ben Bernanke, announced that the Fed was planning to purchase 300 billion in bonds from the markets. While this “quantitative easing” of the environment seems to be needed for the economy, it is also accompanied by mounting worries over high inflation after the recession ends. The analysis below may help explain the limited effects of the interest rate rule, and the possible success of the quantitative easing strategy of the Fed.
5.2.1 Interest rate rule
We assume that the interbank rate will be lowered by 1% in response to a 1% shock to the
10 20 30
collateral effectiveness, and so
D
R . To characterize the persistent interest rate cut in 1reaction to the prolonged crisis, the persistence of the interbank rate,
30.95, is specified. However, compared with Fig. 4, the interest rate cut does not result in much difference under the interest rate smoothing rule, but causes larger declines in consumption and output as well as slight rises in interest rates. The interest rate increases lower the discounted value of the marginal product of capital and thus reduce the value of capital, compared with Fig. 4. The decline in investment worsens the downturn in output and consumption.This peculiar result may be caused by the relative magnitude of the interest rate rule’s responses to the past interest rate and current inflation and output gap. While the 1% cut in the interest rate is not large enough, the interest rate responds more to the current output gap and inflation rate. Because the output rebounds to the normal level in only two quarters, the zero output gap prevents further interest rate cuts. This result is quite robust even under an unrealistic 5% quarterly cut in the interest rate to the shocks or inflation targeting rule with the assumption that
110 and
2 . The calibrations of the 0 same policies under the10 20 30
shock to the monitoring in the banking sector also generate the same results.
The crucial determinant seems to be the persistence of the shocks. If the persistence of the shock is lowered to 0.7, instead of 0.95 as in previous cases, a 1% interest rate cut is enough to help the economy recover regardless of whether it is interest rate smoothing, as shown in Fig. 8. Both consumption and output rise slightly above the normal level, accompanied by an increase in investment.
This implies that the interest rate rule is not effective unless the credit market efficiency is quickly restored, which in turn depends on the confidence of the banking sector in the production of loans. In Fig. 7, we can see that although the banking sector hires more people to monitor the loan making and the value of the capital goods increases, lower consumption implies that the outstanding loans remain at a low level. The lower interest rate is not able to stimulate the banking employment or restore the amount of collateral required to make the loans at the level before the shock hits. However, Fig. 8
10 20 30
rapidly recover. Therefore, the effectiveness of the interest rate rule strongly relies on the
reestablishment of the credit market is crucial for the economy to recover but not the other way round.
5.2.2 Expansionary monetary base
While the interest rate rule does not seem to be effective in the face of the crisis, the expansionary base money seems to work much better. With the persistence of the high-powered money growth equaling 0.9 and responding to a 1% financial shock by a 1% increase in the growth rate of the monetary base, Fig. 9 shows that the consumption and output can grow without experiencing the initial recession. The 1% increase in the base money will cause the interbank interest rate to fall by 1.14% in each quarter, which is equivalent to 4.8% per annum. This decline does not seem to be unrealistic because the Fed has lowered the Federal funds rate by 5% within just one year.17 The essential reason for the effectiveness of the monetary expansion is the quantitative easing environment. While a larger high-powered money growth induces a greater demand for consumption followed by higher deposit and
17 Before the crisis, GM (2007) pointed out that the 1.2% quarterly drop in the interbank interest rate was unrealistic.
loan demand, the interest rates are lowered. A lower interest rate stimulates investments and the capital stock, and is helpful in restoring the quantity of loans to a higher level.
In particular, from the balance sheet of the bank,
H
, we can see that the bank reserves riseL D
with the expansion in high-powered money. Deposits are increased accordingly which can facilitate consumption. This mechanism is absent from the conventional literature on monetary policy, which neglects the banking sector, as well as from the studies on credit channels without considering money.This channel is also absent in an economy with an interest rate rule which crucially depends on the interest rate elasticity of consumption and investments.
6 Conclusion
This paper investigates quantitatively the macroeconomic implications of monetary and fiscal policies under the current financial crisis based on the model in GM (2007). By using the DSGE model with the banking sector, we can successfully characterize the financial crisis that was initiated in the credit markets in terms of the shock to the loan production of banks and see how the shock impacts other sectors within the economy. With money and banking, this model allows for the endogenous determination of various interest rates and the EFP.
The effects of policies are examined quantitatively by means of calibrations based on US data where the crisis started. The calibration results show that expansionary government expenditure can successfully help the economy recover, but the timing and size of the policy’s implementation may matter. When characterized by a 3% increase in government spending in response to a 1% shock to the effectiveness of collateral with an AR(1) coefficient of 0.6, this policy will enable the economy to accumulate enough wealth and recover within four quarters, though by sacrificing initial consumption.
crowding-out effect of fiscal policy due to rising interest rates will be attenuated. Therefore, a model without distinct interest rates may overstate the crowding-out effects and the effects of the fiscal policy may be stronger than was thought. While the credit channel literature focuses on the examination of monetary policies, the implications of credit channels for fiscal policy have not been noted before.
The effects of monetary policy also strongly rely on the way the policy is applied. A 1% expansion in the growth rate of base money in response to a unit shock to the effective collateral can successfully stimulate the economic recovery. The increase in the monetary base can raise the demand for consumption as well as deposits, thereby boosting the economic growth. However, with the high persistence of the financial shock of 0.9, as in all other cases, a 4% p.a. interbank rate reduction by the Fed will fail to pull the economy out of the recession, no matter how persistent the interest rate rule is or how large the initial cut is (interest rate smoothing or inflation targeting). It will only be effective if the financial shock is less persistent. That is, the effectiveness of the collateral needs to be restored soon and thus the banking sector will be able or will be willing to provide enough loans for consumption. This finding coincides with the fading effects of the interest rate rule since the financial crisis broke out which has led to an emphasis on “quantitative easing” policy or massive government spending. It shows that it is the effectiveness of the interest rate relies on the credit market restoration, but not the other way round.
However, the policies in this model are relatively simple: the government expenditure focuses on goods without contributing to productivity or consumption directly in the economy. There will be differences if the emphasis is on productive spending or consumables. The way to finance the government expenditure also matters. Whether the funds for the spending are raised by issuing bonds, money or by taxes will result in different interest rate movements and the EFP, and these will have different effects on the economy.
References
Bernanke, B. S. and A. S. Blinder (1988) “Credit, Money and Aggregate Demand” American Economic Review, 78(2), 435-439.
Bernanke, B. S. and M. Gertler (1989) “Agency Costs, Net Worth, and Business Fluctuations” American Economic Review, 79, 14-31.
Bernanke, B. S. and M. Gertler (1995) “Inside the Black Box: The Credit Channel of Monetary Policy
Bernanke, B. S. and M. Gertler (1995) “Inside the Black Box: The Credit Channel of Monetary Policy