본고는 비용 및 수요 충격 이외에 주택수요 및 금융시장 충격을 고려할 수 있는 뉴케인지언DSGE 모형을 구축하고 정책손실함수와 선형 정책금리 조정함수에 금융안정 관련 지표 변수를추가할 때 경제 충격별 정책효과를 모의실험을 통해 평가하여 보았다. 모의실험 결과 첫째, 모든경제적 충격에 대하여 정책금리가 물가 및 생산 갭 이외에 신용량이나 신용위험 프리미엄에도반응할 경우 물가, 생산 및 신용량 변동성의 가중합으로 정의되는 정책손실이 감소하는 것으로나타났다. 다만 이는 물가변동성의 확대를 대가로 얻어진다. 둘째, 정책금리 결정에 금융안정을고려할 때 비용 충격의 경우 정책손실 감소 효과가 가장 큰 것으로 나타났다. 셋째, 선형 정책금리반응 함수에 금융안정 관련 변수로 신용위험 프리미엄 보다는 신용량을 고려하는 것이 정책효과가우월한 것으로 나타났다.

After the experience of 2008 global financial crisis that had beenamplified by the financial sector, the central bank is mandated the financialstability in addition to the price stability in the name of macro-prudentialpolicy. Despite the mandated obligation of financial stability, however, thecentral bank of the Korean economy has very restrictive policy tools formacro-prudential policy. Liquidity provision as a lender of last resort againstthe banking crisis is an ex-post policy measure, rather than a preemptiveone that can be used for a preventive purpose. The recently discussedrepresentative macro-prudential policy tools such as the loan to value (LTV)ratio, bank levy, or counter-cyclical buffer belong to financial supervisoryauthority. This means close policy coordination between monetary andfinancial supervisory authority is required. But the establishment of suchcoordination is yet doubtful. As a result, the central bank has no choicebut to use the short term interest rate as a macro-prudential policy toolin addition to as a monetary policy tool, at least for the time being. On these circumstances, this study investigates the effects of shortterm interest rate adjustments when they are performed to pursue boththe monetary and macro-prudential policy purposes. The model used inthis paper is basically a new Keynesian DSGE model that is augmented witha financial sector and real estate sector following Bernanke, Gertler, Gilchrist(1999) and Iacoviello (2005) respectively. The monetary policy is in thefunctional form of Taylor rule type that is augmented with financial variablesuch as the credit amount or the credit risk premium. The policy is performedin an optimal way, however, by determining the coefficients of policy functionsso that the central bank``s quadratic policy objective function augmentedwith financial stability variable is minimized. In this sense, the policyimplication in this paper is normative rather than positive. The policyexperiments are fulfilled for the separate four shocks of demand, supply,housing, and financial ones.The four policy implementation methods are experimented in this study. The first is the traditional monetary policy that does not consider themacro-prudential policy at all. In the second policy implementation, thefinancial variable is included not in the central bank``s objective functionbut in the interest rate policy function. This experiments the policy thatconsiders the financial factor to the extent that the factor affects traditionalmonetary policy objectives of price and real sector stability. The third isthe case that the credit amount is used as a financial variable in the interestrate policy function. The fourth is the case that the credit risk premiumis used as a financial variable in the interest rate policy function. The experimental results can be summarized as follows. First, thepolicy loss defined by the weighted average variations of price, output,and credit reduces when the policy interest rate responds to financialvariables. It costs the price stability instead however. Second, the interestrate policy that considers financial stability is most effective to the costshock. This was especially true when the interest rate responds to the creditamount rather than the credit risk premium. Third, as a financial stabilityindicator in the interest rate policy rule, using credit amount turns outto be superior rather than using credit risk premium.