The role of monetary policy on house
prices in California
Does monetary policy significantly affect house prices in
California?
BSc Thesis Economics and Business Name: Emma Penthum Student number: 10545344 Supervisor: Dr. C. G. F. van der Kwaak
1
Abstract
The effect of monetary policy on house prices during the housing bubble has been researched on the United States as a whole before. However, none of this researches focus on the
different impact of monetary policy on house prices among separated states. In this thesis, the role of monetary policy on house in California, one of the most affected states by the housing bubble, is investigated. The monetary transmission mechanism is used to explain the channels through which monetary policy affects house prices. Linear regression analysis is thereafter used to conclude that the effect of monetary policy on house prices is significant in
California.
Keywords: Monetary policy, house prices, housing bubble, monetary transmission mechanism.
Statement of Originality
This document is written by Emma Penthum who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
2
Table of Contents
Introduction 3
1. Economic Theory 5
1.1. The Housing Bubble 5
1.2. Basics of the Monetary Transmission Mechanism 7
1.2.1. User cost of capital 8
1.2.2. Expected appreciation of house prices 8
1.2.3. Housing supply 10
1.2.4. Credit channel 10
1.2.5. Wealth 11
1.2.6. Summary 12
1.3. Monetary Policy, Financial Stability and House Prices in the United States 13
2. The Discussion on the Housing Bubble 15
3. Data analysis 16
3.1 Methodology 16
3.2. Variable Analysis 18
3.2.1. The housing market 18
3.2.2. Federal funds rate 20
3.2.3. Mortgage rate 21
3.2.4. Consumer price index 22
3.2.5. Personal income 23 3.2.6. Unemployment rate 24 3.2.7. Dummy variable 25 3.3 Findings 25 4. Conclusion 28 References 29
3
Introduction
Before 2006, house prices in the United States started rising tremendously and peaked early 2006. In 2006, this so-called housing bubble seemed to have burst and house prices started to decrease with an absolute low measured in 2012. Some economists argue that the credit crisis resulting from the collapse on the house market largely contributed to the recession that started in December 2007 (Holt, 2009). After the burst of the bubble, the value of homes decreased. Because of the high debt financing of homes, financial institutions faced delinquencies of mortgage payments. Also, the value of securities that were related to the house prices devaluated. This caused many of these institutions to get in trouble. Moreover, areas where house prices had been falling after a period of rapid increases in house prices faced a higher increase in the number of foreclosures (Board of Governors of the Federal Reserve, 2008) These are negative effects resulting from the housing bubble.
During a period of rising house prices, according to theory, the pace of house price
increases can be lowered by increasing the interest rate. However, this tends to be at the cost of economic growth (Williams, 2015). This is a problem of monetary policy. A second problem of monetary policy are the shocks and the long lags of transmission to the real economy (Bernanke, 2010).
In current debates among central banks is discussed whether or not monetary policy
should be used to maintain financial stability (Willams, 2015). Many economists argue that stabilization of output and the inflation rate should be maintained by monetary policy
(Mishkin, 1995). Although house prices represent only one channel through which monetary policy is affects stability, it is an important channel to investigate. The reason for this are the negative effects housing bubbles can have on the economy and the problems that come with monetary policy as described above.
One of the states in the U.S. that was affected most by the housing bubble is
California. In California, the house price index increased up to 23.3% per year in 2005. On the contrary, the largest annual increase in the house price index for the United States as a whole was 11.3% per year in 2005. After the burst, house prices in California decreased up to 19.6% per year. Moreover, 1.2 million families lost their homes between 2008 and 2011 in California (Stein et al., 2011).
Many researchers invested timein investigating the impact of monetary policy on
house prices in the United States as a whole during the housing bubble. Many of these previous researches conclude that the effect of monetary policy on house prices is small or
4
insignificant. However, the United States is a large country, where effects can differ among states. The question, therefore, is whether or not the monetary policy of the Central Bank of the United States, the Federal Reserve, had a different effect among states. Hence, in this thesis, the effect of monetary policy, in the form of the interest rate, on house prices in California, one of the most affected states by the housing bubble, is investigated to determine if monetary policy might had a greater effect on this state separately than on the U.S. as a whole. The following question will be investigated: Does monetary policy significantly affect house prices in California?
To answer the research question, a linear regression analysis is done for California.
The outcomes for the coefficients of the federal funds rate for California will be discussed and used to form a conclusion on the research question. These findings will be compared with the existing theory on the relationship between monetary policy and the house prices in the United States.
This research will be conducted in two phases. The first phase of the research will be
based on literature and economic theory. During this phase, the theoretic relationship between monetary policy and the house prices will be described. In the second phase of this research, empirical evidence on the relationship between monetary policy and house prices will be investigated.
In the first section of the first chapter, the housing bubble is explained. The second
section gives an extensive description of the channels of the monetary transmission
mechanism, through which monetary policy affects the economy and the house prices. The last section of this chapter outlines the role of monetary policy in financial stability and during the house bubble. The second chapter provides a summary on the discussion that exists around the role of monetary policy on the house prices during the bubble in the United States. Chapter three provides the findings of the empirical research. In the first section, the methodology is described, which is followed up by the variable analysis and the results from the regression analysis in respectively the second and third section. In the fourth chapter, a conclusion on this research is provided. The conclusion is followed up by the references.
5
1. Economic theory
This chapter will describe the existing theory on this topic. In the first section of this chapter, a general housing bubble is defined, the U.S. housing bubble is described, and some of the suggested causes are named. The second section outlines the basics of the monetary
transmission mechanism. This economic model describes how the Federal Reserve Bank affects the housing market and therefore house prices by the interest rate. In the third and final section of this chapter, monetary policy is defined. Also, the importance of monetary policy and its contribution to financial stability and house price stability within the United States is explained.
1.1. The housing bubble
A housing bubble is a type of asset bubble. This section will describe what an asset bubble, and therefore a housing bubble, is and how it arises. Following, the characteristics of a bubble are described. Finally, the most recent U.S. housing bubble is shortly described and some of the suggested causes of the bubble are summed up.
Typically, asset bubbles are defined as the difference between the fundamental values
and the market values of the assets (Tirole, 1985). Therefore, economists generally use the term “bubble” to describe that the price of an asset has risen above the fundamental value (Kiselev and Ryzhik, 2010). Value investors are specialized in finding these divergent asset prices in order to invest in the undervalued assets. On the contrary, short sellers, investors that sell assets they borrowed from a broker in order to profit from an anticipated decrease in asset price, look for overvalued assets to make a profit (Scherbina, 2013). Eventually, the market will notice the mispricing on the asset market, and a price correction will cause the prices to sharply drop or rise. However, it is often hard to identify these bubbles, because of the difficulty in separating between price changes arising from the underlying asset and those based on “irrational exuberance” (Ahearne, 2005). This term was used first in 1996 by Alan Greenspan, former chairman of the Federal Reserve of the Unites States, and is now often used to describe a situation of increased speculation that drives up asset prices (Shiller, 2005).
There are five phases of a bubble identified in the book by Kindleberger (1978) and in
the work of Hyman Minsky (1972, 1982). The first stage is displacement. In this phase, investors notice changes in the economy or obtain a new belief about the economy and, therefore, start investing in the industry that has profit opportunities according to those beliefs. The second stage is the boom. Prices are slowly rising as a result of the displacement
6
phase. Next, prices start rising more as more investors notice the profit opportunities. This establishes the boom phase. Investors may not be aware of the development of the bubble. The third phase is the phase of euphoria. Asset prices peak, and everyone is aware of the profit opportunities. Speculation increases as non-investors start participating, as
Kindleberger quotes: “there is nothing so disturbing to one’s well-being and judgment as to see a friend get rich” (Kindleberger and Aliber, 2005). This phase tends to be short-term. The fourth stage is the crisis phase, where investors start to sell. Therefore, prices start to fall. This causes the stock or house prices to crash and the bubble bursts. Investors start to panic. The final and fifth stage is revulsion. Investors become appalled by the events, and asset prices reach their low point (Kindleberger, 1978).
Early 2000s, the pace of the increase in house prices in the United States accelerated.
This was the development of the housing bubble. However, there is still an ongoing
discussion on when the housing bubble did actually start (Buyn, 2010). According to Robert Shiller, the housing boom started in 1995. This conclusion was made based on the Case-Shiller 10-city home price index where house prices rose between 1995 and 2006 (2010). However, the pace of acceleration started to increase in the early 2000s. Also, the correction of this housing bubble was much more abrupt than the development of the bubble and comprised a smaller time period (Buyn, 2010).
Many economists have researched this phenomenon and discussed the causes of the
bubble on the U.S. housing market. A housing bubble can be caused by multiple factors. House prices are like other asset prices determined by a market of demand and supply. Therefore, an increase in house prices suggests that there is an increase in demand or a
decrease in supply of houses. According to Ahearne et al. a housing boom is usually preceded by expansionary monetary policy, but followed by contractionary monetary policy when authorities notice that house prices are increasing, before house prices peak. However, this is only a rule of thumb (2005).
In the 2009 article by Sowell, he emphasized the role of governments in the
development of this housing bubble. According to Sowell, the local governments that restricted land supplies saw a larger increase in house prices in their region, compared to regions where the governments were less restrictive (2009). On the other hand, Krugman blamed the unregulated shadow banking system, where traditional banking functions are performed outside of the regular banking system, for the development of the housing bubble. This so-called shadow banking system got highly leveraged during the housing bubble, and deleveraged after its burst, contributing to the credit crisis (2009).
7
Byun argued that the smaller amount of requirements to obtain a mortgage, which
increases risk on such loans, contributed to the housing bubble. The mortgage-backed security, also called MBS, is a type of asset-backed security, whose income payments and value are derived from the underlying mortgages. Investors in these mortgage-backed securities are basically lending money to home owners, and therefore have a claim on the principal and interest payments done by the mortgage holders. These mortgage-backed securities were increasingly used to finance the housing loans and, because they increased ease of obtaining a mortgage, also contributed to the housing bubble (2010).
According to Holt, mortgage lending was eased before and during the period when
house prices rose, because of the high securitization of these mortgages, and because lenders were screened less (2009). Securitization of the mortgages is the practice where residential mortgages are pooled together and small pieces of the pool are sold as mortgage-backed securities, in this case, to investors (Holt, 2009). Moreover, Brueckner et al. name the increased subprime mortgage lending, lending to people who may have difficulty to repay their mortgage, as a cause of an increase in housing demand, and therefore the increasing house prices (2012).
Four factors can be identified as contributing to the housing bubble in the United
States according to Gwartney et al. (2008). These are the increased debt-to-income ratio for households, the increased ease of obtaining a mortgage, the increased leverage of investment banks and lastly, the low short-term interest rate policy of the Federal Reserve (Gwartney et al., 2008). Holt described four factors that look similar to those listed above. According to him, the primary causes of the housing bubble are the irrational exuberance, the increased ease of obtaining a mortgage and the low mortgage and short-term interest rates (2009).
Concludingly, Bernanke states repeatedly that expansionary monetary policy of the
Federal Reserve is named as the cause of the bubble in house prices in the Unites States (2010). The low interest rate is the cause that will be researched further throughout this thesis. In the third section, the impact of the interest rates will be explained.
1.2. Basics of the Monetary Transmission Mechanism
Monetary policy is an important aspect of central banks, but the effects are sometimes unexpected and difficult to measure. To describe how monetary policy affects the economy, the monetary transmission mechanism is used. This mechanism describes the effects on the economy through different direct and indirect channels (Mishkin, 1995). The channels of the monetary transmission mechanism affecting the house prices are explained next. Channels of
8
the monetary transmission mechanism that affect other variables than house prices are disregarded.
Expansionary or contractionary monetary policy, in the form of lowering or raising
the interest rates, has multiple effects on the housing market and therefore on the economy (Miskin, 1995). The effects on house prices and housing demand can be described by
different channels which are divided among direct and indirect channels. The most important direct channels affecting the economy and house prices after monetary policy shocks are the user cost of capital, expectations of future house-price movements, and the housing supply. The indirect channels affecting the economy and house prices are the credit channel and wealth (Wadud et al., 2012). As follows, these direct and indirect channels will be explained.
1.2.1. User cost of capital
Standard neoclassical models view the user cost of capital as an important determinant on the demand for capital. Since houses are a form of capital, the user cost of capital is also an important determinant on the demand for housing (Boivin et al., 2010). The user cost of capital (uc), is a direct channel that is composed out of different components that represent the yearly costs of owning a house (Gutiérrez González, 2005). These components are the purchase price of new housing capital (ph), the interest rate (i) which is adjusted for the marginal tax rate (t), the expected rate of appreciation of house prices (πhe), and the
depreciation rate for house prices (δ) (Miskin, 2007). These variables lead to the following formula for the user cost of capital:
uc = ph[(1 – t)i – πhe + δ].
Expansionary monetary policy, a decrease of the interest rate, lowers the user cost of capital, everything else equal (Case & Shiller, 2003). A lower user cost of capital will lead to a higher demand for capital, and therefore a higher demand for residential capital. In turn, this
increases the demand for housing and the house prices (Wadud et al., 2010). On the contrary, contractionary monetary policy causes the user cost of capital to increase, everything else equal. This leads to a decrease in demand for residential capital, and therefore to a decrease in demand for housing and house prices (Wadud et al., 2010).
1.2.2. Expected appreciation of house prices
The expected appreciation of house prices is a direct channel, which is also part of the user cost of capital. As provided in the equation on the user cost of capital, expectations on the appreciation of house prices affect the user cost of capital and therefore the housing demand
9
(Case and Shiller, 2003). The formula for the user cost of capital can be rewritten as: uc = ph[(1 – t)i – πe – (πhe – πe) +δ]
where πe is the expected rate of general inflation and πhe the expected appreciation of house
prices. The term πhe – πe represents the expected real rate appreciation of house prices
(Mishkin, 2007).
According to the theory on the user cost of capital as described above, an increase in
the interest rate leads to an increase in the user cost of capital and a decrease in house prices. In turn, this increase in the interest rate can, therefore, lower the expected appreciation of house prices. Everything else equal, this increases the user cost of capital even more, and can lead to a lower demand of residential capital, and therefore housing. Lastly, this lower
demand decreases house prices. On the other hand, a decrease in the interest rate leads to a decrease in the user cost of capital, a higher demand for residential capital and therefore housing, and finally to higher house prices. In this case, the decrease in the interest rate can cause the expected appreciation of house prices to increase, which in turn lowers the user cost of capital even more. Concluding, this can lead to an even higher demand for residential capital, and therefore housing, and a higher increase in house prices (Mishkin, 2007).
Moreover, it is important to note that according to theory one should recognize that
housing prices are not only determined by the construction costs, but also by the land where they are built on. If house prices were only determined by construction costs, one can expect that the term πhe – πe will not fluctuate much, since general inflation determines the inflation
on construction costs. However, the recognition that prices are also determined by the land where they are built on can cause fluctuations in the expected appreciation of house prices. This argument can be amplified by the 2007 research by Davis and Heathcote where they found that between 1975 and 2006 the real price of residential land in the U.S. increased by 270 percent, while the increase of the real price of construction costs was only 33 percent. Two factors are named for those fluctuations in the price of residential land. First, in most areas, there are restrictions on the use of land and the number and size of new homes on this land. Therefore, the supply of new homes is limited. Secondly, although the U.S. has empty land available for new houses, some of these land is located on places where most people do not want to live. This is also a land supply restriction (Mishkin, 2007).
Moreover, housing purchases can be seen as an investment vehicle, for example,
when a house is bought for the sole reason of reselling it in order to make a profit. If an appreciation in house prices is expected, the housing demand can be raised because of these investments (Ahearne et al., 2005). This raises house prices even more. However, if the
10
expectations of an appreciation are proven wrong, many of these residential investments return back to the market. This means that investors resell the houses they purchased as an investment in order to make new investments, which may be in the non-housing market. This increases the housing supply and in turn decreases house prices (Ahearne et al., 2005).
1.2.3. Housing supply
Housing supply is a direct channel in the transmission mechanism. Because of the short time horizon of housing construction, the short-term interest rate is a relevant measure of cost. A higher interest rate causes an increase in construction costs by increasing borrowing costs to finance construction materials and therefore causes a reduction in housing supply (Mishkin, 2007). The rise in construction costs is imputable to increased rental costs for equipment needed and the reduction in supply of building materials caused by a fall investment, when assuming these firms invest partly or entirely with debt (Wadud et al., 2012). On the contrary, a lower interest rate causes a decrease in construction cost, by lowering the cost to finance construction materials, and therefore causes an increase in housing supply (Mishkin, 2007). Concluding, in the case of expansionary monetary policy and therefore a decrease in the interest rate, construction costs decrease and the housing supply increases. A higher supply decreases house prices. This is, therefore, a counter-effect of the two direct channels described above. In those two channels house prices increase caused by an increase in housing demand, according to the theory, when the interest rate decreases.
1.2.4. Credit channel
The credit channel is an indirect channel that indirectly affects house prices by directly
affecting housing demand.The credit channel is composed of both the balance-sheet channel
and the bank-lending channel.
As follows, the effects of the credit channel through the balance-sheet channel are
explained. An important issue in the credit market is the existence of asymmetric information. Lenders are hesitant in lending when there is an uncertainty if they will be repaid. Worthy collateral, such as houses, reduces this risk and increases lending, everything else equal. As a result, when house prices increase, the amount of collateral increases. This can improve the amount of credit available to households. This concept is referred to as mortgage equity withdrawal (Mishkin, 2007). Also, most households face credit constraints. This means that their spending is dependent on their income and expenses. In the case of a short term interest rate change, this change will affect the current cash flows of households.
11
This leads to two possible channels. The first channel suggests that housing demand can be affected by the nominal interest rate. A higher interest rate causes a reduction in the current cash flows of credit-constrained households that borrow, since their current payments increase. This reduces the size of the mortgage households can afford and therefore causes a lower housing demand. On the other hand, a lower interest rate causes the contrary, and can cause a higher housing demand.
Moreover, according to the neoclassical framework, only changes in long-term rates
should affect housing demand, and it does not matter if the homeowner owns a variable-rate or a fixed-rate mortgage (Mishkin, 2007). However, the second balance-sheet channel suggests that it is relevant whether the homeowners have variable or long term interest rates, if they face credit constraints. The reason for this is that, according to this channel, not only changes in the long-term interest rate affect housing, but also changes in the short-term interest rate. For example, if a large part of the mortgages is financed with a variable-rate mortgage, and the short-term interest rate changes, this can have an effect on housing demand even if the long-term interest rate remains unchanged. The reason for is that credit
constrained households will have lower interest payments and a higher cash flow or higher interest payments and a lower cash flow when the interest rate decreases or increases, respectively (Mishkin, 2007).
As follows, the credit channel effects trough the bank-lending channel are explained.
This channel essentially affects the lending institutions. Monetary policy can affect the amount of money and loans, from the central bank, available to commercial banks and institutions. Also, through reserve requirements, monetary policy can affect the amount of reserves financial institutions must hold. They, therefore, can impact the amount of money available for lending to, for example, households for home financing. This can affect housing demand (Iacovielllo and Minetti, 2007). However, in this thesis, only the impact of monetary policy through the interest rate is investigated. Therefore, this channel is not relevant.
Summarizing, the credit channel suggests that an increase in the interest rate reduces
the cash flows of credit constraint households and therefore decreases housing demand and that a decrease in the interest rate increases the cash flows of credit constraint households and, therefore, increases housing demand.
1.2.5. Wealth
Wealth is an indirect channel in the transmission mechanism that is directly affected by house prices. In turn, wealth directly affects the housing demand (Wadud et al., 2012). As stipulated
12
above, lower interest rates will result in higher housing demand and therefore raise house prices. This leads to an increase in total wealth, which thereafter results in higher
consumption and aggregate demand (Mishkin, 2007). This statement is supported by the standard applications of the life-cycle hypothesis theory by Modigliani and Brumberg (1954) whom state that any increase in wealth will increase household consumption. Frederic
Mishkin asserts that the standard life-cycle wealth effects that are operating through house prices are therefore an important element in the monetary transmission mechanism (Miskin, 2012). The ability to consume more, can lead to a higher housing demand.
However, there is still an active discussion on the wealth channel. In extreme cases
one could even argue that household consumption could be reduced if potential house buyers believe they have to save more to be able to pay the higher housing prices (Miskin, 2007). The change in household consumption also depends on the marginal propensity to consume (Guiliodori, 2005). Elbourne suggest that because of the high amount of household wealth in real estate, more attention should be paid to the wealth effects that are caused by the
increasing house prices (2008).
1.2.6. Summary
I now present a graphic illustration of the channels described above in Figure 1. In this thesis, the effect of expansionary monetary policy, in the form of a decrease in the interest rate, on house prices in the U.S. is investigated. Therefore, this figure represents the effects of the monetary transmission mechanism on house prices when the interest rate decreases.
13
1.3. Monetary Policy, Financial Stability and House Prices in the United States
The Federal Reserve can affect the economy of the United States by monetary policy. There are three main tools to conduct monetary policy, which are through the reserve requirements, the discount rate and open market operations. Reserve requirements are the fraction of deposits that banks must keep as reserves at the central bank. The discount rate is the interest rate that the Federal Reserve charges on short-term loans from depository institutions. Open market operations are used to increase or decrease liquidity the amount of currency in circulation. The main reason why this so-called OMOs are used is to manipulate the money supply or the short-term interest rate. OMOs are the most often used tool of the Federal Reserve (Federal Reserve Bank of St. Louis). The effect of the short-term interest rate, represented by the federal funds rate, on house prices in California is investigated in this thesis. Other tools that are used less are the interest on required reserve balances and excess balances, overnight reserve repurchase agreement facility and the term deposit facility (Federal Reserve Bank).
Using these three main tools, the Federal Reserve has to achieve their goals. The goals of the Federal Reserve are stable prices, moderate long-term interest rates and to promote maximum employment (Federal Reserve Bank). The Federal Reserve cannot choose among these goals, but they make a decision on how much weight is attributed to each goal. In the view of housing bubbles, some central bankers argue that although targeting asset prices is not a main goal, the Federal Reserve should be aware of inflation and output effects when maintaining financial stability (Ahearne et al., 2005). The use of the different tools to maintain the different goals is important, since all policies affect the economy differently through the channels of the monetary transmission mechanism, which are explained in section 2 of this chapter.
In this part, the monetary policy in the United States during the development of the
housing bubble between 2002 and 2006 is briefly discussed. Del Negro and Otrok state that they unsurprisingly found that the deviations from the policy rule, the Taylor rule, were more loose than tight (2006). The Taylor rule is a rule of thumb that describes what the target short-term nominal interest rate should be when considering the long-term interest rate, current inflation, the inflation gap and the output gap (Taylor, 1993). Before 2002, the U.S. economy suffered from the ending of the dot-com boom and also from the terroristic attack on 9/11. Stock prices decreased. A moderate recession was present between March and November 2001. After this recession, the target for the federal funds rate was lowered, in order to boost the economy (Bernanke, 2010). Bernanke argues that this expansionary
14
monetary policy was driven by two main factors. First, the recovery of the recession was slow and there was still a high unemployment rate. Second, inflation was below the inflation target, against one of the goals of the Federal Reserve (2010). These issues caused the Federal Reserve to carry out aggressive monetary policy, in the form of low interest rates. After the third quarter of 2004, monetary policy tightened again.
Monetary policy deals with two issues when related to house prices. The first problem
is how to deal with the uncertainty around the monetary transmission mechanism. The second problem is how monetary policy can best respond to fluctuations and bubbles in house prices (Mishkin, 2007). House prices are not the only channel through which monetary policy affects financial stability, but it is a very important one (Williams, 2015). A first reason to focus on financial stability through house prices is mentioned in the introduction, and is the burst of the housing bubble in the United States in 2006. This burst had large effects on the credit market and afterwards on the entire economy. The main reasons for the large effects on the credit market are the high leverage of financial institutions and the expansion of mortgage debt, funded by mortgage-backed securities, in the run-up to the burst of the bubble in the U.S. This increase in debt occurred for a large part in the shadow banking system and these mortgage-backed securities were in turn not secured by the federal government or the originating banks (Wachter, 2014). Also, a rise in real estate finance is visible (Williams, 2015). This increases the importance of financial stability on the housing market, because the risk of real estate finance increases when house prices undergo large fluctuations. However, when a housing bubble arises, it is not straightforward how the bubble can be kept under control. Moreover, it is not straightforward how the effects on the economy of rapidly decreasing house prices, in case of a burst, can be limited (Ahearne et al., 2005).
Also, the importance of the role of financial supervision in monitoring the exposure of
banks in the case of a price decrease is emphasized (Ahearne et al., 2005). In his 2010 speech at the Annual Meeting of the American Economic Association in Georgia, Bernanke,
chairman of the Board of Governors of the Federal Reserve System at the time, stated that economists who hold the Central Bank responsible for the housing bubble argue for a greater role of monetary policy in preventing such bubbles. However, their opponents argue that the bubble was not caused by monetary policy, and that this is neither a good tool to control and stabilize house prices (2010). In the same speech, he therefore suggest that the best response against the housing bubble would have been to use regulatory policy, not monetary policy (Bernanke, 2010).
15
2. The discussion on the housing bubble
The opinion on whether or not monetary policy affected the housing bubble differs among economists. Most argue that the housing bubble is caused by the expansionary monetary policy of the Federal Reserve Bank, but this view has opponents.
Anna J. Schwartz (2009) blames the Federal Reserve Bank for their expansive
monetary policy. Also, John Taylor (2009) argues that the Federal reserve maintained the interest rate too low during the critical period of the rising house prices, which caused a rapid inflation of housing prices.
Furthermore, Del Negro and Otrok conclude that although the Federal Reserve
performed more expansionary deviation than contractionary deviation from the policy rule, that this deviation was not significant. Also, the overall effect of monetary policy on house prices was small in comparison to the total increase in house prices, according to their research. They, however, do not conclude that the U.S. government is not responsible for the housing boom (2006). Also, Shiller argues in his article that the housing bubble was not entirely or partly caused by the expansionary monetary policy. The several reasons named by Shiller are that the period of the low interest rates was much shorter than the period of rising house prices, the mortgage rate did not adjust to the monetary policy until after the boom had already begun, and that there are many other economic variables that caused the housing price boom (2009).
In another research, Dokko et al. argue that theory suggests that loose monetary
policy affected the house prices, but that there is no statistical evidence for this (2011). According to Ben Bernarke, the direct linkages between monetary policy and the rise in house prices is weak. This is caused by the long lags between implementing policy and the moment when the economy is actually affected (2010).
Some of these articles suggest that the effect of monetary policy on the U.S. house
prices is very small or even insignificant. However, none of this researches focusses on the differences between states. The United States has many states with different sub-economies. Therefore, the impact of monetary policy on house prices can be researched on the U.S. as a whole, but the effects may differ among states. Consequently, the effect of monetary policy on house prices in only California, one of the most affected states by the housing bubble, is investigated to determine if the effect of monetary policy on house prices is also small or insignificant in a state that is largely affected by the housing bubble.
16
3. Data analysis
In this chapter, the methodology of this research is explained in the first section. This section gives an extensive description of how the empirical research is conducted. Also, the
hypothesis is shown. In the second section, the variables used are explained and are graphically presented. In the third section, the data found during the empirical research is presented and discussed.
3.1 Methodology
To empirically investigate the research question, a linear regression analysis will be done in STATA for California. Next, the coefficients of monetary policy on housing prices obtained by the regression will be discussed. The period that will be researched contains data from October 1st 1975 until October 1st 2015. The frequency of the data is quarterly, and therefore this research contains 160 data points.
First, to do a regression, the model has to be determined. In this research, the change
in house prices are the dependent variable. The independent variables that will be used to explain the change in house prices are the previous changes in house prices, the change in consumer price index, change in personal income, the federal funds rate (short-term interest rate) of the two previous quarters, the mortgage rate (long-term interest rate) of the previous quarter, the unemployment rate, a dummy variable for the housing bubble and two interaction variables of the housing bubble combined with the two lagged variables for the change in the house price index. There is also an error term added to the model. The following regression model is constructed as presented in equation 1:
∆ HPIt = β0 + β1∆HPIt-1 + β2∆HPIt-2 + β3it-1 + β4it-2+ β5mgrt-1 + β6∆CPIt + β7∆PIt +
β8ut + β9Dt + β10∆HPIt-1*Dt + β11∆HPIt-2*Dt + εt (1)
In words, the percentage change with respect to the previous quarter of the house price index on time t, ∆ HPIt, is explained by a constant, β0, the percentage change in the house price
index of the previous quarter and two quarters earlier, ∆HPIt-1 and ∆HPIt-2 respectively, the
federal funds rate from one quarter and two quarters earlier, it-1 and it-2 respectively, the
mortgage rate of the previous quarter, mgrt-1, the percentage change of the consumer price
index with respect to the consumer price index of the previous quarter, ∆CPIt, the percentage
17
the unemployment rate, ut, a housing bubble dummy, Dt, two interaction variables between
the dummy and the lagged change in house prices indexes, HPIt-1*Dt and ∆HPIt-2*Dt
respectively, and an error term, εt. The choice of variables, what the variable precisely entails
and their sources will be explained in the next section, section 3.2.
Up next, the coefficients of the federal funds rate on house prices in California will be
analyzed. Because of the lags that monetary policy deals with, this model includes the federal funds rate of t-1 and t-2. This means that the federal funds rate of the previous quarter and two
quarters earlier is used. The following hypothesis will be tested:
H0: β3 = β4 = 0
H1: β3 β4 0
This means that the null hypothesis is that both the coefficients for the federal funds rate on the house prices are equal to zero. Assuming normality, this is a two-sided t-test. Also, the coefficient for the dummy variable that indicates the housing bubble will be highlighted. The expectation is that this coefficients will not be equal to zero, since house prices in California were largely affected by the housing bubble.
Also, it is important to check the robustness of the results. This is a common check for
empirical studies to examine how the so-called core regression behaves when certain regressors are added or removed from the model (Lu and White, 2014). To investigate the robustness of the regression presented in equation 1, four extra regression analyses are performed. In the second regression, one lag on the federal funds rate is eliminated. In the third regression, the lags for the change in house price index are eliminated. In the fourth regression, the dummy variable for the housing bubble is removed, and therefore, also the interaction variables between the dummy and the lags for the change in the house price index. In the fifth and final regression, the change in the consumer price index, the change in
personal income and the unemployment rate are removed. Therefore, this regression depends on the lags of the change in house price index, the lags of the federal funds rate, the lag of the mortgage rate, the dummy and the interaction variables of the dummy only. The equations that will be regressed are presented next.
∆HPIt = β0 + β1∆HPIt-1 + β2∆HPIt-2 + β3it-1+ β4mgrt-1 + β5∆CPIt + β6∆PIt + β7ut +
18
∆HPIt = β0 + β1it-1 + β2it-2+ β3mgrt-1 + β4∆CPIt + β5∆PIt + β6ut + β7Dt + β8∆HPIt-1*Dt +
β9∆HPIt-2*Dt + εt (3)
∆HPIt = β0 + β1∆HPIt-1 + β2∆HPIt-2 + β3it-1 + β4it-2+ β5mgrt-1 + β6∆CPIt + β7∆PIt +
β8ut + εt (4)
∆HPIt = β0 + β1∆HPIt-1 + β2∆HPIt-2 + β3it-1 + β4it-2+ β5mgrt-1 + β6Dt + β7∆HPIt-1*Dt +
β8∆HPIt-2*Dt + εt (5)
However, the main regression being investigated presented in equation 1 deals with some difficulties. Monetary policy is very unpredictable and comes with long lags. In a linear regression, these lags are partly ignored. Therefore, most economists use the vector autoregression model to investigate the relationship between monetary policy and house prices. Such a VAR model accounts for all the individual lags of dependent variables and also takes into account the relationships between dependent variables that can exist when
determining the model. When using a linear regression model instead of a VAR model, the results of the regression can be biased, because the independent variables may be correlated with the error term. Broadly said, this can occur when something related to the house prices is also related to one of the independent variables, but is not included in the model. This is called simultaneity. It can also be caused by a measurement error, since lags are not included. This problem is called the omitted variable bias.
3.2 Variable Analysis
This section briefly discusses the variables and their sources that are used in this thesis.
3.2.1. The housing market
In the period from 1974 to 2002 the residential investment in the United States was 4.5% of GDP, which rose after 2002 to 6.25% at the end of 2005 (Dokko et al., 2011). This increases the importance of house prices.
During the most recent housing bubble in the U.S., house prices rose above average in
California. This is visible in Figure 2, where the all-transactions house price index in California is compared with the all-transactions house price index of the United States as a whole.
19
Figure 2: House Price Index. Source: U.S. Federal Housing Finance Agency.
Figure 3 amplifies that the effect of the housing bubble was more extreme in California than in the United States on average. In 2005, the largest annual increase in the house price index was measured. House prices rose by 23.3% that year in California. On the contrary, house prices rose by only 11.3% in the United States as a whole. Moreover, after the collapse of the housing bubble, house prices decreased rapidly in California. Between 2007 and 2008, the house price index decreased by 19.6%. For the U.S. as a whole, the largest decreased was measured between 2008 and 2009, where the house price index decreased by 5.5%. This emphasizes the importance of investigating California separately.
Figure 3: Annual Change in House Price Index in Percentages. Source: U.S. Federal Housing Finance Agency.
0,00 100,00 200,00 300,00 400,00 500,00 600,00 700,00 1975-01-01 1988-09-09 2002-05-19 California United States -25 -20 -15 -10 -05 00 05 10 15 20 25 30 1976-01-01 1989-09-09 2003-05-19 California United States
20
In this thesis, the house prices and, therefore, also the previous house prices variables
are determined by the house price index. The variables used are the change of the house price index with respect to the previous period. The reason that the change of the house price index is used instead of the house price index, is because the house price index is a non-stationary variable. Using the percentage change with respect to the previous period makes this variable stationary. The all-transactions house price index is be used. Data is from the U.S. Federal Housing Finance Agency obtained by the Federal Reserve Bank of St. Louis.
3.2.2. Federal funds rate
The federal funds rate can be manipulated through open market operations, which is one of the tools of monetary policy. The federal funds rate represents the short-term nominal interest rate. As explained in the third section of the first chapter, monetary policy loosened early 2001, caused by a moderate recession. The following figure, figure 4, shows the federal funds rate over the entire period that is being investigated. As visible, the federal funds rate has fluctuated over time. Large decreases are most often the effect of a recession, where the Federal Reserve wanted to stimulate the economy and investment by lowering the federal funds rate. The recessions in the early 1980s, the early 1990s and the early 2000s recessions are all followed by a visible decrease in the federal funds rate.
Figure 5 zooms in on the period of the most recent housing bubble. Within one year, from December 2000 and December 2001, the Federal Reserve lowered their target federal funds rate from 6.5% to 1.5% (Dokko et al., 2009, p. 1).This figure clearly shows the expansionary
0,00 2,00 4,00 6,00 8,00 10,00 12,00 14,00 16,00 18,00 20,00 1975-01-01 1988-09-09 2002-05-19
Federal funds rate
21
monetary policy starting after the dot com bubble in 2000 up to 2004 when the housing bubble was developing. In 2004, there’s an increase in the federal funds rate, up to 2008,
when the financial crisis started and the federal funds rate came to an absolute low.
Figure 5: The Federal Funds Rate During the Housing Bubble. Source: Board of Governors of the Federal Reserve.
The federal funds rate is the interest rate at which depository institutions lend balances at the Federal Reserve to other depository institutions overnight. This interest rate is used as a measure of monetary policy, since it affects short term interests rates. In this thesis, two lagged variables for the federal funds rate are used, the federal funds rate from the previous quarter and two quarters earlier. Data is from the Board of Governors of the Federal Reserve System and is obtained from the Federal Reserve Bank of St. Louis.
3.2.3. Mortgage rate
In this thesis, the mortgage rate is also included in the regression. For the mortgage rate, the 30-year fixed-rate mortgage rate is used. Data is from Freddie Mac, obtained by the Federal Reserve Bank of St. Louis.
There is also a decrease visible in early 2000s in the mortgage rates, the long-term
interest rate to finance home mortgages, as it was for the federal funds rate. This is shown in figure 6. In 2003, the rate for mortgages was 5.3%, which was a decrease of 3% with respect to the 2000. Between 2004 and 2008, the federal funds rate started rising again. However, there is not a large increase visible in the mortgage rate between 2004 and 2008 as visible in
0,00 1,00 2,00 3,00 4,00 5,00 6,00 7,00 2000-01-01 2002-09-27 2005-06-23 2008-03-19 2010-12-14
22
Figure 5. For the mortgage rate, the largest decrease is visible in the early 1980s, when the U.S. was in a recession.
Figure 6: The 30-year fixed mortgage rate in percentages between 2000 and 2011.
3.2.4. Consumer price index
The consumer price index measures changes in price levels. The CPI is obtained for all urban consumers on all items the United States as a whole. Data is from the U.S. Bureau of Labor Statistics and is obtained from the Federal Reserve Bank of St. Louis. The change in the consumer price index represent inflation in this regression. The percentage change with respect to the previous period is used and these are seasonally adjusted. The percentage change per quarter of the consumer price index is used to change the non-stationary variable, the consumer price index, into a stationary variable. This is clearly visible in the following two figures, Figure 7 (non-stationary variable) and Figure 8 (stationary variable).
The change in consumer price index is included as a measurement of inflation.
Although there are many more variables affecting house prices, theory suggest that there is a positive relationship between inflation and house prices (Anari and Kolari, 2002).
0,00 2,00 4,00 6,00 8,00 10,00 12,00 14,00 16,00 18,00 20,00 1975-01-01 1988-09-09 2002-05-19 Mortgage rate
23
Figure 7: Consumer Price Index. Source: U.S. Bureau of Labor Statistics.
Figure 8: Percentage Change of the Consumer Price Index. Source: U.S. Bureau of Labor Statistics.
3.2.5. Personal income
Personal income is measured as total personal income and is obtained separately for California. The percentage change with respect to the previous period is used and these are seasonally adjusted. Data is from the US Bureau of Economic Analysis obtained by the Federal Reserve Bank of St. Louis. As well, the percentage change per quarter of the personal income is used to change the non-stationary variable, the consumer price index, into a
stationary variable. This is clearly visible in the following two figures, Figure 9
(non-0,000 50,000 100,000 150,000 200,000 250,000 300,000 1976-01-01 1989-09-09 2003-05-19
Consumer Price Index
-2,0 -1,5 -1,0 -0,5 0,0 0,5 1,0 1,5 2,0 1976-01-01 1989-09-09 2003-05-19
24 stationary) and Figure 10 (stationary).
Personal income is included since it affects wealth. A higher income means a higher
wealth, which in turn increases investment. Therefore, an increase in personal income can increase housing investment and demand, which in turn increases house prices (Mishkin, 2007).
Figure 9: Total Personal Income. Source: U.S. Bureau of Economic Analysis.
Figure 10: Percentage Change in Total Personal Income. Source: U.S. Bureau of Economic Analysis.
3.2.6. Unemployment
Unemployment is measured as a percentage, the unemployment rate. The unemployment rate measures the percentage of the labor force that is unemployment, but is seeking for
employment and willing to work. Data is from the US Bureau of Labor Statistics and is
0 500000000 1000000000 1500000000 2000000000 2500000000 1976-01-01 1989-09-09 2003-05-19
Total Personal Income
-6,0 -4,0 -2,0 0,0 2,0 4,0 6,0 1976-01-01 1989-09-09 2003-05-19
25 obtained from the Federal Reserve Bank of St. Louis.
Unemployment is included in this regression, since it affects wealth. In turn, a change in wealth can affect the amount of investment and, therefore, also investment in houses. A change in housing demand affect house prices (Mishkin, 2007). The unemployment rate over the period being researched is presented in Figure 11. Worth mentioning is the high increase in the unemployment rate after the burst of the bubble, which indicates the impact of the following financial crisis.
Figure 11: The Unemployment Rate in California. Source: U.S. Bureau of Labor Statistics.
3.2.7. Dummy
As explained, this regression analysis includes a dummy variable. The reason for this is to account for the housing bubble that was present early 2000s. Excluding the housing bubble may give biased outcomes. Therefore, the dummy variable is 1 when the data is from a period between the first quarter of 2002 up to and including the first quarter of 2006. Otherwise, the dummy is 0. Moreover, to appropriately account for the housing bubble, there are also two interaction variables added to the model. The dummy is combined with the two lagged variables for the change in house price index into two interaction variables.
3.3. Findings
This section will provide the findings of this research. The output is presented and the outcomes are discussed. Following, the results from the regression analyses are presented in Table 1. 0,0 2,0 4,0 6,0 8,0 10,0 12,0 14,0 1976-01-01 1989-09-09 2003-05-19 Unemployment Rate
26
Table 1: Regression analyses
∆ HPIt (1) (2) (3) (4) (5) ∆HPIt-1 0.9024*** (0.0896) 0.9119*** (0.0925) 0.8036*** (0.0800) 0.9024*** (0.0876) ∆HPIt-2 -0.353 (0.0947) -0.0354 (0.0978) 0.0411 (0.0832) -0.0311 (0.0889)
Federal funds rate at t-1 0.2010
(0.1484) -0.1174 0.1558 (0.2497) 0.1681 (0.1490) 0.2124* (0.1233)
Federal funds rate at t-2 -0.4454***
(0.1357) -0.5694** (0.2283) -0.4223*** (0.1398) -0.4159*** (0.1311) Mortgage rate at t-1 0.2916** (0.1275) 0.1359 (0.1221) 0.4524** (0.2138) 0.3038** (0.1243) 0.2401** (0.1032)
∆Consumer price index -0.5809
(0.2060) 0.0377 (0.2105) 0.4545 (0.3362) -0.0867 (0.2101) ∆Personal income -0.0250 (0.1064) -0.0262 (0.1098) 0.5630*** (0.1682) -0.0190 (0.1092) Unemployment rate -0.0660 (0.0802) -0.0370 (0.0823) -0.4646*** (0.1249) -0.0722 (0.0785) Dummy 2.7956*** (0.8841) 2.7645*** (0.9125) 1.9471 (1.4866) 2.9275*** (0.8609) ∆HPIt-1*Dummy -0.6478*** (0.1940) -0.6236*** (0.2001) 0.0773 (0.2899) -0.6473*** (0.1905) ∆HPIt-2*Dummy 0.0033 (0.1917) 0.0245 (0.1977) -0.0477 (0.2817) 0.0025 (0.1866) Constant -0.3596 (0.6134) -0.0779 (0.6270) 1.9198* (0.9915) -0.7546 (0.4692) N 160 160 160 160 160 R2 0.7723 0.7558 0.3449 0.7497 0.7706 Probability F > 0 0.0000 0.0000 0.0000 0.0000 0.0000 * significant at 10% ** significant at 5% *** significant at 1%
The regression of equation 1 is the main regression being researched. The regressions of equation 2, 3, 4, and 5 are used to see how robust the results of the main regression are.
First of all, the coefficient for the one quarter lagged variable of the percentage
change in house price index is 0.9024, and is significant at a level of 1% in regression (1). The extra regressions verify the this outcome with similar coefficients and also significance levels of 1%. A possible explanation for a positive coefficient is that economic agents see a positive percentage change in house prices, and therefore expect house prices to appreciate. In these adaptive expectations, the expectation of house price appreciation depends on the
27 appreciation of house prices in the previous period.
The most important variables in this thesis are the federal funds rate at t-1 and t-2. The
coefficient for the one quarter lagged variable for the federal funds rate is insignificant in all the regression. However, the coefficients for the two quarters lagged variable for the federal funds rate is -0.4454 in the main regression, and significant at a level of 1%. This outcome is confirmed by the other regressions, where the coefficient is also negative and significant at 1% or 5%. This outcome is in line with what the theory discussed in this thesis suggests. A possible explanation why only the two quarters lagged variable is significant, is because a one quarter lag may be too short for the housing demand and house prices to adjust to a change in the federal funds rate.
The main regression shows a significant outcome at a 5% level for the mortgage rate.
The coefficient found is 0.2916. Except for regression (2), the outcome for the coefficient of the mortgage rate is also positive and significant at a level of 5% in the other regressions. The outcome is, however, not what was expected. According to this outcome, a rise in the
mortgage rate would increase the house prices. This contradicts the theory. A possible explanation for this is the omitted variable bias that this regression deals with.
The coefficient for the dummy variable is 2.7956 and is significant at 1%. Two out of
the three extra regressions confirm this positive coefficient for the dummy of the housing bubble at a level of 1%. This outcome is in line with the theory. During the housing bubble, house prices in California extraordinary. The dummy accounts for this.
The final significant outcome that the main regression shows is for the coefficient of
the interaction variable between the one quarter lagged change in the house price index and the dummy variable for the housing bubble. The coefficient for this is -0.6478, and is significant at 1%. Regression (2) and (4) show similar outcomes for this variable. This outcome contradicts the outcomes of these two variables separately, which both showed a positive and significant coefficient. However, this may be just a moderation of those outcomes.
28
4. Conclusion
This last chapter contains the final conclusion on the research question: Does monetary policy significantly impact house prices in California?
This thesis provided an extensive description on how monetary policy affects house
prices through the monetary transmission mechanism. Three direct channels and two indirect channels, the user cost of capital, the expected appreciation of house prices, the housing supply, the credit channel and wealth, are explained. Summarizing, four of these channels suggest that there is a positive effect on house prices when the federal funds rate decreases. Only the housing supply channel contradicts this theory.
Researchers in previous literature are inconclusive on the effect of monetary policy on
house prices as discussed in chapter 2. Some economists found small impacts, others, on the other hand, found that there is no significant effect of monetary policy on house prices. However, previous researches only focused on the United States as whole. The United States is a large country, where effects can differ among states. Since California was one of the most affected states by the housing bubble, this thesis focused on the impact of monetary policy on house prices in only California.
The main regression analysis showed that there is no significant effect of the one
quarter lagged variable of the federal funds rate. However, the two quarters lagged variable did have a significant negative impact on house prices. Also, there is a significant large positive coefficient for the dummy for the housing bubble, which indicates that there was a large impact on house prices during that period. The outcomes found in the main regression seemed to be robust and were confirmed by the extra regressions. The final conclusion on the research question is that monetary policy did affect the house prices in California, although the impact of monetary policy on house prices is lagged by half a year.
However, one must be cautious with these results, because of the simultaneity and
29
References
Ahearne, A. G., Ammer, J., Doyle, B. M., Kole, S. K., Martin, R. F. (2005). House Prices and Monetary Policy: A Cross-Country Study. Board of Governors of the Federal Reserve
System, No. 841, pp. 1-64.
Anari, A., Kolari, J. (2002). House Prices and Inflation. Real Estate Economics, Vol. 30(1), pp. 67-84.
Assenmacher-Wesche, K., Gerlach, S. (2008). Financial Structure and the Impact of Monetary Policy on Asset Prices. Working Paper 2008-16, Swiss National Bank. Bernanke, B.S. (2010) Speech at the Annual Meeting of the American Economic Association,
Atlanta, Georgia. Board of Governors of the Federal Reserve System.
Bjørnland, H. C., Jacobsen, D. H. (2010). The Role of House Prices in the Monetary Policy Transmission Mechanism in Small Open Economies. Journal of Financial Stability,
Vol. 6(4), pp. 218-229.
Board of Governors of the Federal Reserve (2008). Monetary Policy Report to the Congress, submitted pursuant to section 2B of the Federal Reserve Act.
Boivin, J., Kiley, M. T., & Mishkin, F. S. (2010). How Has the Monetary Transmission Mechanism Evolved Over Time? National Bureau of Economic Research, No. 15879, Massachusetts, Cambridge.
Brueckner, J.K., Calem, P.S., Nakamura, L.I. (2012). Subprime Mortgages and the Housing Bubble. Journal of Urban Economics, Vol. 71(2), pp. 230-244.
Byun, K.J. (2010). The U.S. Housing Bubble and Bust: Impacts on Employment. Monthly
Labor Review, December 2010.
Case, K.E., and Shiller, R.J. (2003). Is There a Bubble in the Housing Market? Brookings
Papers on Economic Activity, Vol. 2, pp. 299-362.
Capozzo, D.R., Hendershott, P. H., Mack, C., Mayer, C.J. (2002). Determinants of Real House Price Dynamics. National Bureau of Economic Research, No. 9262, Massachusetts, Cambridge.
Davis, M.A., Heathcote, J. (2007). The Price and Quantity of Residential Land in the United States. Journal of Monetary Economics, Vol. 54(8), pp. 2595-2620.
Del Negro, M., Otrok, C. (2007), 99 Luftballons: Monetary Policy and the House Price Boom Across States. Journal of Monetary Economics, Vol. 54(7), pp. 1962-1985.
Dokko, J., Doyle, B.M., Kiley, M.T., Kim, J., Sherlund, S., Sim, J., Van Den Heuvel, S. (2011). Monetary Policy and the Global Housing Bubble. Economic Policy, Issue 66,
30 pp. 233-283.
Elbourne, A. (2008). The UK Housing Market and the Monetary Policy Transmission Mechanism: An SVAR Approach. Journal of Housing Economics, Vol. 17(1), pp. 65-87.
Federal Reserve Bank. Policy Tools.
https://www.federalreserve.gov/monetarypolicy/policytools.htm/
Federal Reserve Bank. What are the Federal Reserve’s objectives in conducting monetary
policy? https://www.federalreserve.gov/faqs/money_12848.htm
Federal Reserve Bank of St. Louis. How Does Monetary Policy Work?
https://www.stlouisfed.org/in-plain-english/how-monetary-policy-works
Friedman, M. (1968). The Role of Monetary Policy. The American Economic Review, Vol.
58(1), pp. 1-17.
Giuliodori, M. (2005). The Role of House Prices in the Monetary Transmission Mechanism across European Countries. Scottish Journal of Political Economy, Vol. 52(4), pp. 519-543.
Gwartney, J., Macpherson, D., Sobel, R., Stroup, R. (2008). The Crash of 2008: Cause and Aftermath. Cencage.com.
http://www.cengage.com/economics/book_content/0324580185_gwartney/pdfs/Speci
al_Topic_Crash_of_2008.pdf
Holt, J. (2009). Summary of the Primary Causes of the Housing Bubble and the Resulting Credit Crisis: a non-technical paper. The Journal of Business Inquiry, Vol. 8(1), pp. 120-129.
Iacoviello, M., Minetti, R. (2008). The Credit Channel of Monetary Policy: Evidence From the Housing Market. Journal of Macroeconomics, Vol. 30(1), pp. 69-96.
Kindleberger, C.P. (1978). Manias, Panics and Crashes: A History of Financial Crises. (First edition). New York: Basic Books.
Kindleberger, C.P., Aliber, R.Z. (2005). Manias, Panics and Crashes: A History of Financial
Crises. (Fifth edition). Hoboken, New Jersey: John Wiley and Sons, Inc.
Kiselev, A., Ryzhik, L. (2010) A Simple Model for Asset Price Bubble Formation and Collapse. Cornell University.
Krugman, P. (2009). The Return of Depression Economics and the Crisis of 2008. (First edition). New York: W. W. Norton.
Lu, X., White, H. (2014). Robustness Chekcs and Robustness Tests in Applied Economics.
31
McDonald, J.F., Stokes, H. H. (2013). Monetary Policy and the Housing Bubble. The Journal
of Real Estate Finance and Economics, Vol. 46(3), pp. 437-451.
Minsky, H. P. (1972) Financial Instability Revisited: The Economics of Disaster. Reappraisal
of the Federal Reserve Discount Mechanism, pp. 97-136.
Minsky, H. P. (1982). The Financial Instability Hypothesis: Capitalistic Processes and the Behavior of the Economy. Financial Crises: Theory, History, and Policy, edited by Charles P. Kindleberger, Jean-Paul Laffargue, 12-29. Cambridge: Cambridge University Press.
Mishkin, F.S. (1995). Symposium of the Monetary Transmission Mechanism. The Journal of
Economic Perspectives, Vol. 9(4), pp. 3-10.
Mishkin, F.S. (2007). Housing and the Monetary Transmission Mechanism. Board of
Governors of the Federal Reserve System (U.S.), No. 2007-40, pp. 359-413.
Romer,C.D., Romer, D.H. (2013). The Missing Transmission Mechanism in the Monetary Explanation of the Great Depression. American Economic Review, Vol. 103(3), pp. 66-72.
Rigobon, R., Sack, B. (2004). The Impact of Monetary Policy on Asset Prices. Journal of
Monetary Economics, Vol. 51(8), pp. 1553-1576.
Scherbina, A. (2013). Asset Price Bubbles: A Selective Survey. International Monetary
Fund.
Shiller, R. J. (2005). Irrational Exuberance. (First edition). Princeton: Princeton University Press.
Shiller, R. J. (2009). Unlearned Lessons from the Housing Bubble. The Economists’ Voice,
Vol. 6(7), pp.
Shiller, R. J. (2010). Don’t Bet the Farm on the Housing Recovery. New York Times,
Economic View. 05-06-2016.
http://www.nytimes.com/2010/04/11/business/economy/11view.html
Simo-Kengne, B., Miller,S., Gupta, R., Balcilar, M. (2016). Evoluation of the Monetary Transmission Mechanism in the US: The Role of Asset Returns. The Journal of Real
Estate Finance and Economics, Vol. 52(3), pp. 226-243.
Sowell, T. 2009. The Housing Boom and Bust. (First edition). New York: Basic Books. Stein, K., Bedrossian, K., Hopson, R. (2011). California’s Fight Against Foreclosures Goes
National. Race, Poverty & the Environment, Vol. 18(1), pp. 70-73.
Taylor, J. (1993). Discretion versus Policy Rules in Practice. Carnegie-Rochester Conference
32
Wachter, S. (2014). The Market Structure of Securitisation and the US Housing Bubble.
National Institute Economic Review, Vol. 230(1), pp. 34-44.
Wadud, I.K.M., Omar, H.M.N., Huson, J.A.A. (2012). Monetary Policy and the Housing Market in Australia. Journal of Policy Modeling, Vol. 34(6), pp. 849-863.
Williams, J.C. (2015). Measuring Monetary Policy’s Effect on House Prices. FRBSF
