The effectiveness of the post-crisis measures in the United States and the Euro-zone

(1)

The effectiveness of the post-crisis measures in the United States and

the Euro-zone

University of Amsterdam

Bachelor Thesis Economics and Business

By: Mike Trott

Student number: 10264620

Supervisior: C.G.F. van der Kwaak

Date: 7-5-2015

Faculty of Economics and Business Track: Economics and Finance Field: Macroeconomics

Abstract

After the unfolding of the Great Recession economies have witnessed a decrease in economic activity and increase in unemployment. Since then the U.S. and the E.U. have responded with a variety of measures. In the U.S., Treasury implemented TARP, the U.S. congress the ARRA and the Federal Reserve implemented Quantitative Easing. In the E.U., the ECB implemented the Enhanced Credit Support and the European Commission the EERP. This paper examines the different measures with the use of

economic literature and assess the performance with the use of reference values like the NAIRU, output gap and real economic growth rates. The analysis shows that in terms of recovery the U.S. is performing better than the E.U. with rising output gaps, lower unemployment and higher real GDP growth rates. However, a limitation of this paper is that there is non-conclusive evidence from empirical literature on the effectiveness of the measures implying that there is more difficulty in comparing the effectiveness of the measures taken.

(2)

Chapter 1: Introduction

In 2008 the world was hit by the Great Recession and its effects are still visible today; economies are experiencing high unemployment rates and low economic growth. The primary cause for the

recession was the increase in leverage by banks (Brunnermeier, 2009). Leverage is a term used for banks or firms that utilize borrowed money to amplify the potential profit or loss. Banks used debt to finance their investments in risky assets with a higher payoff. This can be seen as excessive risk taking by banks. A tool used to increase leverage was securitization. Securitization ultimately led to lower mortgage rates and lower interest rates on other types of loans (Brunnermeier, 2009). The lower mortgage rates

allowed more consumers to get mortgages and to buy houses which increased the demand for housing. The increase in demand led to higher prices in the housing market. This ultimately created a bubble in the housing market (Brunnermeier, 2009). Since there were low checks on creditworthiness the mortgage defaults started increasing causing the bubble to burst. Due to the increase in mortgage defaults credit rating agencies began noticing the actual risk implied by securitization and a run on the securitization system occurred. This ultimately led to the collapse of the financial funds and bank insolvency (Mishkin F. S., 2011). This subject will be discussed in more detail later.

Since the beginning of 2007 central banks and governments have responded with a variety of measures. In September 2008 the U.S. launched the TARP program in which it bought equity stakes in financial institutions and the European Central Bank implemented the Enhanced Credit Support which tried to improve bank lending to consumers. It should be noted that these are just two examples and more measures will be discussed later.

The effectiveness of all the measures together has not yet been assessed by other literature. A lot of research has been done to assess the effectiveness of individual measures like Quantitative easing by Joyce, Miles, Scott & Vayanos (2012), fiscal policy by Ilzetzki, Mendoza, & Végh (2013), TARP by Nguyen & Enemoto (2009) or the effectiveness of the American Recovery and Reinvestment Act by Feyrer and Sacerdote (2011). However there is no direct comparison between all the direct post crisis measures taken by the E.U. and the U.S.

The contribution of this thesis is to make a direct comparison between the U.S. and the E.U. with respect to the post crisis measures and to make an assessment on which economy is recovering faster than the other. A limitation of this assessment is that we only look at certain performance

(5)

This leads to the following research question: “Comparing the post-crisis measures of the U.S.

and the E.U., to what extent do they differ and what do they have in common. Also if there is a performance difference then which performed better?”

To answer the research question the paper will start in chapter 2 by building a theoretical framework in which several measures will be explained with the use of the IS-LM model. With the use of the IS-LM model the effects of fiscal stimuli and monetary policy will be explained separately. After that the interaction between monetary and fiscal policy, the commercial bank balance sheet and finally Quantitative easing and its effects will be explained.

In chapter 3 the pre-crisis environment and the events leading to the crisis will be explained to elaborate on the crisis factors the economies were facing. After explaining the factors leading to the crisis the measures taken by the economies will be discussed as well as the goals of both central banks.

In chapter 5 the methodology and the source of the data as well as a short analysis will be given. The data will be analyzed using variables like NAIRU, real GDP growth rates and output gap. This will give us an indication of the performance of the economies after the measures. Chapter 6 will discuss the empirical results giving a better indication on the effectiveness of the measures. Chapter 7 is the conclusion in which the main findings of the paper will be recapitulated.

(6)

Chapter 2: Theoretical framework

In order to understand the effects of the different measures taken by the governments and central banks there needs to be a theoretical foundation. This can be done by using the IS-LM model. The IS-LM model will help to clarify the way in which monetary policy and fiscal policy interact and how they affect economic activity. First the IS and the LM curve are established. Then there will be an

elaboration on how the models reacts to different shocks. The second thing to be done is to see how the models react to the measures taken by the governments.

2.1 ISLM model

2.1.1 IS curve

The IS curve represents the various combinations between the interest rate and income in which the goods market is in equilibrium. The goods market is in equilibrium when 𝑌 = 𝐶 + 𝐼 + 𝐺 + 𝐸𝑋 − 𝐼𝑀, where Y represents national income, C represents national consumption, I represents investment, G represents government expenditure, EX represents Export and IM represents Import (Gärtner, 2009). To develop the full model of the IS curve, the individual components of the equation need more explanation.

The first element examined is the consumption function. Consumption does not only depend on the current income but also on the expected future income. This makes sense since knowing that you will have more money in the future will make you spend more now and if you have more income now it is also quite likely that you will spend more now, since the present value of income is higher (Mankiw, 2010). The consumption function can therefore be explained by the following equation: 𝐶 = 𝑐1𝑌 + 𝑐2𝑌𝑒. However, a more simplified consumption function provided by Gärtner (2009) will be used, in which the expected future income will be kept constant and equal to 0. This gives the following

consumption function 𝐶 = 𝑐𝑌 where c is the marginal propensity to consume. The marginal propensity to consume measures the effect of the increase in current income on the total consumption (Gärtner, 2009). This implies that if we observe a propensity to consume of 0.5 a rise in national income by 1 leads to a rise in consumption of 0.5. Another important factor in consumption is taxes (Gärtner, 2009). Taxes have a negative relation with consumption since a higher tax rate implies a lower disposable income and thus lower consumption. If taxes are included in the consumption curve, the consumption function would be: 𝐶 = (1 − 𝑡)𝑐𝑌 (Gärtner, 2009).

(7)

The investment function consists of two major factors, namely the expected future income and the interest rate (Gärtner, 2009). The interest rate is negatively related to the total investment function, since a higher interest rate implies higher cost of borrowing which decreases the demand for

investment. The investment function is also positively related to the expected future income since a higher future income would mean that there will be excessive cash flow later, which could be invested now (Mankiw, 2010). This gives us the following investment function: 𝐼 = 𝑏1𝑌𝑒− 𝑏2𝑖 (Gärtner, 2009). However to make the analysis simpler we assume that the future income is constant and hence 𝑏1𝑌𝑒will be constant and equal to 𝐼̅. This implies the following investment function: 𝐼 = 𝐼̅ − 𝑏𝑖.

The Export function depends on the world income and the Real exchange rate. Both are positively related to the Export. The real exchange rate, R, is given by R ≡𝐸×𝑃_𝑃𝑤𝑜𝑟𝑙𝑑 (Gärtner, 2009), E represents the nominal exchange rate in domestic currency per foreign currency, 𝑃𝑤𝑜𝑟𝑙𝑑 the price of a bundle of goods in foreign currency and P the cost of the same bundle of goods in the home currency. If the nominal exchange rate depreciates against other currencies, which implies that E rises, thus making the home country’s products relatively cheap for foreigners. This in turn will stimulate export (Gärtner, 2009). Export is also positively related to the world income since a higher world income increases the demand for domestic goods by the rest of the world. The increase in demand means that the import of the world will rise since not all products can be produced or are available in the rest of the world. Since the worlds import is the home countries’ export, this implies that the home countries’ export will rise (Mankiw, 2010). This gives us the following function 𝐸𝑋 = 𝑥1𝑌𝑤𝑜𝑟𝑙𝑑+ 𝑥2𝑅.

The import function is the exact opposite of the export function since the import of the home country is the export of the world and vice versa. This implies the following function 𝐼𝑀 = 𝑚1𝑌 − 𝑚2𝑅.

Substituting the functions, including taxes, and solving for the interest rate gives the IS curve, according to Gärtner (2009): 𝑖 = − (1 − 𝑐 + 𝑐𝑡 + 𝑚1 𝑏 ) 𝑌 + ( 𝑥2+ 𝑚2 𝑏 ) 𝑅 + 𝐼̅ + 𝐺 + 𝑥1𝑌𝑤 𝑏 (1)

As we can see the IS curve is downwards sloping and negatively related to the GDP.

2.1.2 LM curve

The LM curve represents the various combinations between the interest rate and income in which the money supply is equal to the money demand.

The money demand consists of two elements, it can be a medium of exchange and a store of value (Mishkin, Matthews, & Guiliodori, 2013). Money as a medium of exchange has a positive relation with the income, since having more income leads to more expenditures which in turn leads to a higher

(8)

demand for money (Mankiw, 2010). Money as a store of value consists of a precautionary demand for money and a speculative demand for money. Precautionary demand for money is money held for covering unexpected events. Speculative demand for money is money held at times when other assets have higher risks. Both depend primarily on the interest rate, which is negatively related to the money demand (Gärtner, 2009). A higher interest rate implies a higher opportunity cost of holding money which means that consumers will be more likely to store their money into their bank accounts and withdraw the money when they absolutely have to, rather than holding cash on hand (Mankiw, 2010). Another option is that they would invest it. Both lead to a lower money demand. Therefore money demand can be summarized as follows 𝑀𝑑 = 𝑘𝑌 − ℎ𝑖 (Gärtner, 2009).

The money supply is controlled by the central bank, by inducing sales or purchases of

government treasury bills or by buying and selling foreign currency (Mishkin, Matthews, & Guiliodori, 2013). To simplify the analysis the money supply is seen as an external factor and is therefore kept constant, as stated by Gärtner (2009). This can be denoted as 𝑀𝑠 = 𝑀̅.

The LM curve represents the combinations between the interest rate and national income where the money demand equals money supply (Gärtner, 2009). Therefore setting the money demand and money supply equal to each other and solving for the interest rate gives us the following LM curve

function: 𝑖 =𝑘_ℎ𝑌 −1_ℎ𝑀̅ (2)

Figure 1 basic plot of the IS and LM curves. On the y-axis the interest rate is noted and on the x-axis the GDP.

Figure 1 shows the basic plot of the ISLM model with a downward sloping IS curve and an upward sloping LM curve. It should be noted that in the figure above the central bank targets the money

IS LM

(9)

2.1.3 Policy stances and the ISLM model

Now that the IS-LM model has been established, the effects of different measures can be illustrated. The first measure that will be examined is monetary policy. Monetary policy can have two stances, it can either be expansionary or contractionary. Expansionary monetary policy occurs when a central bank expands the money supply by buying government bonds in order to reduce the real interest rate and induce economic growth (Mishkin, Matthews, & Guiliodori, 2013). Contractionary monetary policy occurs when a central bank tightens the money supply to increase the interest rate and to slow down or stop economic growth (Mishkin, Matthews, & Guiliodori, 2013). The effect of

expansionary monetary policy on the IS-LM model, with a constant initial output, can be illustrated as follows; the increase in the money supply shifts the LM curve to the right and thereby lowers the interest rate. In other words the government creates an excess supply of money so that the opportunity cost of money, the interest rate, drops. This will stimulate GDP growth since it will be relatively cheap to borrow money, thus stimulating investment and consumption (Mankiw, 2010). The exact opposite holds for contractionary monetary policy (Mankiw, 2010).

The second measure examined is fiscal policy. Fiscal policy can also be either expansionary or contractionary. Expansionary fiscal policy can be conducted in two ways; increasing government expenditures and lowering taxes. Increasing government expenditure means that the government will invest in infrastructure, office supplies etc. to boost the total production, total income and also the level of employment. The effect of increasing government expenditures can be seen in the IS curve with a rise in G, which implies that the IS curve will shift to the right and thereby increase output. Lowering taxes increases disposable income this will therefore shift the IS curve to the right since a higher disposable income leads to a higher consumption and thus output (Gärtner, 2009).

The effect of expansionary fiscal policy on total income depends on the fiscal multiplier. The fiscal multiplier shows with how much the national income will change, in percentages, when changing government expenditures with 1%. Also, the effect of expansionary fiscal policy on GDP also depends highly on the composition of the government expendtiture (Ilzetzki, Mendoza, & Végh, 2013)

An important factor for the fiscal multiplier is whether or not the Zero Lower Bound (ZLB) is binding or not. The ZLB is reached when the nominal interest rate is at zero percent. When the ZLB is not binding and interest rates are positive the increase in government expenditure will shift the IS curve to IS’ and raise income. See Figure 2. The raise in income will create an excess demand since production cannot keep up with demand. This will raise the price level and thus causing inflation. Since the interest rate is not at the zero lower bound the central bank will intervene by raising the nominal interest rate

(10)

and thus the real interest rate. This will raise savings and lower consumption thus mitigating the effect of expansionary fiscal policy.

Figure 2 ISLM model with non-binding ZLB. On the y-axis the interest rate is noted and on the x-axis the GDP.

Once the ZLB is binding, as seen in figure 3, the government should increase its spending (Eggertsson, 2009). The impulse will shift the IS curve rightwards to IS’ and thereby increasing expected inflation. However, with a binding ZLB there will be no change in the nominal interest rate. It should also be noted that in this case the real interest rate is negative, however the nominal interest rate cannot be negative. When the interest rate is at the ZLB and there is an increase in expected inflation, the real interest rate decreases causing consumption and investment to rise and thus output (Christiano,

Eivhenbaum, & Rebelo, 2011).So in contrast to a non-binding ZLB the binding ZLB does not experience a mitigation of its effects since there is no feedback on the nominal interest rate.

Figure 3 ISLM model with binding ZLB. On the y-axis the interest rate is noted and on the x-axis the GDP. -2% 0% 2% 4% 6% 8% 10% 12% 14% 0,00 1,00 2,00 3,00 4,00 5,00 6,00 7,00 8,00 9,00

IS-LM non binding ZLB

IS LM IS' 0% 2% 4% 6% 8% 10% 12% 14% 0,00 1,00 2,00 3,00 4,00 5,00

ISLM Model with non binding ZLB

(11)

In order to get a better understanding of further measures there must first be a better

understanding of the structure of banks and the way they manage their balance sheet. This will provide insight in the measures taken by governments and central banks.

2.2 Commercial Bank balance sheet and bank recaps

2.2.1 Assets

The balance sheet of a commercial bank implies that the total assets are equal to the liabilities and capital. First the asset side of the balance sheet will be discussed. The asset side consists of reserves, securities, loans and other assets.

When a bank acquires reserves by deposits from consumers, the bank deposits the physical currency at the banks account at the central bank (Mankiw, 2010). The reserves can be divided into two major groups, namely the required reserves and excess reserves. The bank is required to keep a certain amount of the deposits as reserve at the central bank in order to make sure it has enough liquidity to provide for sudden withdrawal of funds, these are called required reserves. Excess reserves are reserves held in excess of the required reserves. The purpose of excess reserves is to provide extra liquidity in case of sudden funds withdrawal and act as a cushion in an environment of uncertainty (Mankiw, 2010). Banks also invest in long and short term securities. Short term securities and bonds are highly liquid assets which can quickly be converted into cash (Mishkin, Matthews, & Guiliodori, 2013). Banks invest in these securities since they can have higher payoffs than when banks stores their money at the central bank (Mankiw, 2010). In essence these securities form a second type of reserves since they can quickly be converted in to cash to provide for the withdrawal of funds.

Banks do not only generate income with the use of securities but also by issuing loans. These include loans to firms, consumers, banks and also mortgages (Mankiw, 2010). Banks borrow at a lower rate from other banks or central bank and they originate loans with a higher interest rate. Since they receive more interest than they have to pay they make a profit (Mankiw, 2010).

2.2.2 Liabilities

The liability side of a banks’ balance sheet consists of sight deposits, time deposits, banks’ deposits and borrowings, debt and other securities, foreign currency deposits and bank capital (Mankiw, 2010).

Sight deposits, also known as demand deposits, represent the account in which a depositor can immediately withdraw or deposit money from or to his account (Mankiw, 2010). An example of a transaction is withdrawing money from a bank account at an ATM or writing checks to other parties. Sight deposits are on the liability side since they represent an obligation of bank, because the bank is

(12)

obliged to pay when funds are required.

Time deposits are accounts in which funds are kept for a fixed amount of time so they cannot be withdrawn on demand (Mankiw, 2010). Debt and other securities represent various loans from the financial market, these include bonds and certificates of deposits. Certificates of deposits are similar to the savings account of a customer (Mishkin, Matthews, & Guiliodori, 2013). However, they have a fixed maturity, a fixed interest rate and can be resold in the financial market.

2.2.3 Capital and Bank Recaps

The final category is the bank capital which is the banks’ net worth. Net worth is the difference between assets and liabilities and represents the residual income to investors (Mankiw, 2010). For example, categories that are listed under capital are shareholders’ equity and retained earnings. Capital acts as a cushion against value changes of its assets. These value changes could lead banks into

insolvency, however the net worth can prevent this since it acts as cushion (Mishkin, Matthews, & Guiliodori, 2013).

In the context of leverage, leverage is the total amount of debt you have to finance your assets. In normal conditions leverage enables you to increase your profit by increasing the liabilities and thus total assets. Now consider a bank with a high leverage ratio and a sudden loss of value of some of its assets. A high leverage ratio implies a smaller amount of equity compared to its debt. If the drop in value is bigger than what the bank has in equity, the bank will be insolvent. This could mean that a small drop in asset value could potentially have a large effect on the financial health of a bank. (Mishkin, Matthews, & Guiliodori, 2013).

Now that the balance sheet has been established basic banking can be explained. When a deposit is made there is an equal increase in the excess reserves. With the increase in excess reserves banks can make long term loans. In essence banks are transforming short term deposits into long term loans, a process called maturity transformation (Mishkin, Matthews, & Guiliodori, 2013).

When the crisis struck commercial banks saw a sudden decrease in bank capital. This led to governments trying to help the banks. Governments provided capital in exchange for preferred stocks. This would raise the banks’ capital and would lead to a less restricted lending standard leading to more loans to firms and consumers, since the cushion is larger. This is called recapitalization of banks and would help banks absorb the losses and improve bank lending.

(13)

2.3 Quantitative easing

The final measure examined is Quantitative Easing (QE). QE is observed when the interest rates are at the zero lower bound and when conventional monetary policy has no effect on the interest rates. The goal of QE is to lower the medium to long-term interest rates and therefore stimulating the

economy. There are three main channels in which QE can affect the economy; the Portfolio Substitution Channel, Signaling channel and the Bank Funding Channel (Joyce, Miles, Scott, & Vayanos, 2012).

2.3.1 The portfolio substitution channel

When the central bank purchases government bonds, it will reduce the supply of privately held bonds which means that the price of government bonds will go up and the yield will go down. Many companies prefer to match the maturities of assets and liabilities. These companies will use a part of the proceeds from the purchases of bonds by the central bank to purchase other assets like corporate bonds. This will restore the maturity mismatch (Joyce, Miles, Scott, & Vayanos, 2012). If a central bank purchases the long dated assets the stock of privately held long-dated asset is reduced, which implies a lower duration risk. Duration risk arises from the fact that as the time to maturity of an asset increases, the sensitivity of the price to interest rate fluctuations also increases. A lower duration risk will lead to lower term premia on all long-dated assets, so the prices of other long-dated risky assets are very likely to rise as well (Joyce, Miles, Scott, & Vayanos, 2012). The rise in prices will generate capital gains for households thus increasing their wealth. This will increase consumption and therefore GDP. This channel is known as the Portfolio Substitution Channel (Joyce, Miles, Scott, & Vayanos, 2012).

2.3.2 The Bank funding Channel

As mentioned before, when commercial banks have relatively low capital they will lend less. When the central bank purchases bonds from the commercial banks it provides the commercial banks with extra cash. The physical money should go on the bank accounts of these consumers and in turn should lead to higher reserves of the banks, but possibly also higher liabilities. Higher reserves should in turn lead to more lending, which would increases investment and consumption and therefore GDP. This however only occurs when the liabilities do not increase as well (Joyce, Miles, Scott, & Vayanos, 2012).

2.3.3 Signaling channel

The signaling effect occurs when the Federal Reserve uses QE as a commitment to keep the interest rates low. If this commitment is credible, the market may see QE as a signal that the Fed will keep the federal funds rate low. This can lower the future federal funds rate and according to the expectations hypothesis this can have an effect on all interest rates since the expectation hypothesis says that the long term interest rate is equal to the mean of the short term interest rates.

(14)

Chapter 3: Subprime crisis

Now that there is a better understanding of the several measures that can be taken by governments and central banks and their effects on output, the start of the current crisis will be

explained to provide more insight on the measures that were actually implemented by governments and central banks.

3.1 Pre-crisis environment

Prior to the subprime crisis the U.S. was experiencing low interest rates due to large capital inflows from abroad (Brunnermeier, 2009). Asian countries pegged their currency to the dollar by buying or selling U.S. securities and the Federal Reserve had adopted an accommodative policy which can be seen in Figure 4 in the period of 2004-2006 (Brunnermeier, 2009). This holds since the federal funds rate is lower than the Taylor rate in the period of 2004-2006. By buying U.S. securities the dollar appreciated against Asian currency, this made Asian products relatively cheap and increased the export of the Asian countries (Brunnermeier, 2009). However, the increased demand for U.S. securities led to rising prices and since yields on securities and their price have a negative relationship, it lowered interest rates on securities. As stated earlier accommodative monetary policy increases the money supply and lowers the interest rate. This means that the opportunity cost of money, the interest rate, has gone down. Figure 4 shows several differently calculated Taylor rates and the federal funds rate. The Taylor rule shows how monetary policy should be conducted when considering macroeconomic factors like inflation and output gap (Taylor, 1993). It represents the interest rate in which there is a stable economy in the short term but still economic growth in the long run. Comparing the Taylor rate to the federal funds rate shows us how well the Fed conducted monetary policy when only considering output gap, inflation target and actual inflation. Figure 4 shows differently calculated Taylor ruse as well as the targeted Federal Funds rate. Figure 4 also shows that the interest rate prior to the crisis was too low according to the Taylor rule. However, the Fed was concerned about deflation and therefore did not increase the interest rate (Mishkin F. S., 2011).

(15)

Figure 4 The Taylor rule and nominal interest rates (Kahn, 2010)

3.2 Development of the crisis

Due to financial innovation, the traditional “originate & hold” banking model in which banks hold loans until they are fully repaid was replaced by the “originate & distribute” banking model. In the originate & distribute model the bank originated loans. From there the originate & distribute model bundled loans, like mortgages, consumer loans etc., to form diversified portfolios (Mishkin, 2011). The diversified portfolios would then be sliced into different tranches in which each tranche had a different risk characteristic. This process is called securitization. The safest tranche was known as the “senior tranche”, which had a low interest rate and low risk since the senior tranche was paid out first. The most risky tranche was the “toxic-waste” tranche which was paid out last but had a high possible payoff. The different tranches would then be transferred to Special Purpose Vehicles (SPV) which sells the assets to investors (Mishkin, 2011).

In essence banks were transferring the risk of default to the investors by selling the asset backed securities. This meant that banks had little incentives to check the credit worthiness of investors

because they were not bearing the risk (Mishkin F. S., 2011). However, shifting the risk to investors allows for lower interest rates on loans and mortgages. Since more investors bought the MBSs the overall demand increased, leading to higher prices of the MBSs and lower interest rates. Also according to Brunnermeier (2009), “Securitization also allowed certain institutional investors to hold assets

(indirectly) that they were previously prevented from holding by capital requirement.” Financial

institutions were allowed to hold assets with a certain credit rating like AAA. Due to financial innovation and securitization there were more securities with a higher rating which meant that they could hold assets that they were previously prohibited to hold (Brunnermeier, 2009).

To increase the profits banks needed to increase the volume of loans. A problem banks were facing was that investors prefer to have short term securities so in order to originate more loans, the long-term loans needed to be converted into short term loans. The solution was to set up a “Special

(16)

Investment Vehicle” which invested in long term assets and issued short term assets known as “Asset Backed Commercial Paper” or ABCP (Mishkin F. S., 2011). They are called asset backed since the income payments are derived and backed up by collateral.

Since the Special Investment Vehicle invested in long term assets and issued short term assets, there was a risk of not being able to pay its short term debt. When investors stop buying the ABCPs, the SIV is in need of funds in order to pay the investors. However, since the long term assets they are selling are relatively illiquid there is a higher risk of not being able to pay the investors (Mishkin F. S., 2011). When the Special Investment Vehicles were unable to pay their debt payments, the SIVs had two options; use the credit line from banks or fire sales. When a SIV runs into liquidity problems the sponsoring bank, which is the bank that set up the SIV, can provide the SIV with liquidity in order to repay its short term debt. This is also known as a credit line (Brunnermeier, 2009). In a fire sale the SIV quickly sells its assets at low prices in order to raise funds. This is not a problem when one SIV opts for a fire sale but when more SIVs do the same, the prices of the ABCPs can decrease. Which can lead to a lower income from the fire sales and thus a lower contribution to the repayment of investors.

Eventually the lower mortgage interest rates and the few checks on creditworthiness led to people buying second and third houses for capital gains. This led to an increase in demand for houses and thus to an increase in house prices. However, the low checks on creditworthiness led to less creditworthy subprime mortgages and loans. This implies a higher risk of default on the loan. So when the interest rates started to increase, mortgages started to default since the low income subprime borrowers could not pay the higher interest rates. When mortgages default the commercial bank repossesses the houses and tries to sell them (Brunnermeier, 2009). Due to the increasing amount of mortgage defaults there was an excess supply of houses causing the house prices to drop which meant that the underlying value of the MBSs dropped as well. However, the subprime mortgage only

accounted for approximately 13% of the total outstanding mortgages but still had an impact (Federal Reserve Bank of San Fransisco, 2009).

Due to the increasing mortgage defaults and the decline in housing prices, credit rating agencies started to take notice that ABSs had a higher risk than initially perceived (Brunnermeier, 2009). Credit rating agencies then started to downgrade the ABSs and investors stopped buying the ABSs since they were more risky. This brought many of the SIVs into liquidity problems since they needed to sell ABSs to

(17)

This meant that the sponsoring banks extended the credit lines to the SIV. The commercial banks also needed liquidity so they stopped lending to other banks, this was amplified by the decrease in trust between commercial banks since they could not tell what the potential losses were at other commercial banks. The fire sales led to excess supply of ABSs and therefore drove the prices down (Brunnermeier, 2009). Since banks also held ABSs in their own portfolio, the asset side on the balance sheets detoriated as well as their capital.

The detoriation of the balance sheet of banks ultimately led to banks filing in for bankruptcy. One of the major investment banks Lehman went bankrupt in 2008. Many investors bought Credit Default Swaps as insurance against the insolvency of Lehman. A CDS is an agreement between two parties in which the buyer of the CDS will be compensated in case of default of the qualified third party. Due to the mortgage defaults the exposure to CDS losses increased, this eventually led to the default of AIG since AIG was the sole writer of CDSs. (Brunnermeier, 2009).

The collapse of AIG was seen as the trigger of the great recession as we know it (Mishkin, 2011). After the collapse the FED responded with the TARP to provide relief. The effects also spread to other countries and nations. For example Europe, since European institutions also held MBSs. After these events the central banks and governments responded with a series of measurements which will be discussed in the next chapter.

Chapter 4: Crisis measures

In this section the goals of the Central Banks will be discussed as well as the measures taken by the Central banks and governments to counteract the crisis.

4.1 Goals of the Central Banks

The main goals of the U.S. Central Bank, the Federal Reserve Bank (Fed), are to promote low unemployment, stable prices and low long term interest rates. By using monetary policy the Fed can achieve the goal of stable prices. Stable prices in the long term will lead to maximum sustainable output growth and employment. Next to these goals the Fed can also affect financial stability. However, this is not a goal on its own. It is a tool to influence economic performance (Federal Reserve Bank, 2004). The main objectives of the European Central bank is to maintain price stability. According to European Central Bank (2006) the benefit of price stability is its contribution to increasing economic welfare and growth.

(18)

4.2 Measures taken by the U.S. Government

4.2.1 Structural measures

The American Recovery and Reinvestment Act

In the period between Q1 2007 and Q2 2009 the quarterly GDP growth rate of the U.S. had decreased by as much as almost 3% (Federal Reserve Bank, 2014). To recover the economy, the U.S. Congress implemented the American Recovery and Reinvestment Act, also known as the ARRA. The ARRA was a fiscal stimulus package designed to counteract the crisis.

The immediate goals of the ARRA were to decrease the unemployment rate and improve economic activity. To achieve these goals the ARRA originally consisted of a $700 billion government impulse but was expanded to $840 billion and is divided between several categories. The categories consisted of Tax benefits, providing Contracts, Grants and Loans and Entitlement Programs

(Congressional Budget Office, 2011).

The Tax benefit category consist of a series of tax cuts for the working-class and new home owners. It included tax benefits for firms investing in energy efficient products. The tax benefits for the working class and investment in eco-friendly products should increase disposable income and stimulate investment and thus increase GDP (Congressional Budget Office, 2014). The tax benefits for new home owners should stimulate investment since the lower taxes make it cheaper to buy houses. This implies an increase in demand for housing. Thus stimulating investment in houses and thus income for the contractors. The Contracts, Grants and Loans (CGL) category consisted of government expenditures in education, transportation, infrastructure, environment, science, housing, health and other programs. The final category of Entitlements comprised of investments for improving medical care, unemployment insurance, family service, energy, economic recovery, housing and agriculture (Congressional Budget Office, 2014). The goal of the CGL category was to provide work for companies by raising government investment. A higher employment means that there are more people with income who can spend more and thereby increase GDP.

The most recent data shows that about 36% of the total budget is allocated to the Tax Benefits category and 32% is allocated to the Contracts, Grants and Loans category (Congressional Budget Office, 2014).

(19)

The Troubled Asset Relief Program

In 2008 U.S. Treasury implemented Emergency Economic Stabilization Act in order to stabilize the financial sector. The act led to the creation of the Troubled Asset Relief Program, also known as TARP. Initially TARP would invest $750 Billion in buying the MBSs from banks (Chodorow-Reich, Feiveson, Liscow, & Woolston, 2012). However, TARP currently consists out of six programs each to stabilize a different part of the economy. The programs include; Auto Industry, Credit Market Programs, Housing, Bank Investment Programs, and Investment in AIG (U.S. Department of Treasury, 2013). However, the main idea behind TARP is to recapitalize banks in order to restore the credit market and therefore not all components of the TARP will be discussed.

AIG was one of the biggest insurance corporations of the U.S. and writer of CDSs. When the crisis struck and companies defaulted AIG had to pay out the CDSs (Mishkin F. S., 2011). If AIG was not able to pay its CDSs then the CDSs would be worthless. So the government rescued AIG because it was “too big to fall” meaning that if AIG filed in bankruptcy there was a big chance that other institutions would fall as well (Mishkin F. S., 2011). The Government therefore provided AIG with a capital injection and debt guarantees, totaling $182 billion, to rescue of AIG from bankruptcy (U.S. Department of Treasury, 2013).

The Auto Industry program was designed to save the American automotive industry. In the U.S. most of the cars are bought on loans and since the start of the crisis consumer loans were hard to get since banks decreased their lending. This meant that the automotive industries sales decreased significantly during the crisis (Statista, 2012). If automotive companies defaulted many suppliers would be affected as well which would lead to the loss of many jobs and more companies defaulting. In order to save the automotive industry the U.S. government provided loan agreements that required the automotive sector to provide a restructuring plan (U.S. Department of Treasury, 2014).

Due to the crisis it was getting increasingly difficult for consumers and firms to get loans from banks. In order to restore bank balance sheets by providing new capital to increase lending, the

government implemented the Credit Market Program. The Credit Market Program consisted out of; the Public-Private investment Program and the Securities Purchase Program (U.S. Department of Treasury, 2014). The goal of the Public-Private Investment Program was to restore the market for mortgage-backed securities by purchasing MBSs from financial institutions using equity capital from private investors, debt and funds from TARP (U.S. Department of Treasury, 2014). This would increase demand for the MBSs and raise their prices. This would help banks to remove the mortgage backed securities from the balance sheet and provide liquidity to banks which they could in turn use to invest, for example

(20)

by lending to households.

To stabilize the banking sector the Bank Investment Programs were implemented. The Bank Investment Program comprises out of five programs. The Asset guarantee program in which the government supported important institutions by agreeing to absorb losses on the distressed ABSs. However, only the Supervisory Capital Assessment Program and the Capital Purchase Program will be discussed, since these are the most important measures for this thesis. The Supervisory Capital

Assessment Program (SCAP) determined the health of the banks by stress-testing them and in addition it provided capital to those institutions who needed it. In a stress test a period of unfavorable economic conditions is simulated to see whether the bank has enough capital to withstand the unfavorable economy. With an unfavorable economy we mean a crisis. By stress testing the banks Treasury wanted to assure investors that banks had enough capital to deal with a significant withdrawal of funds (U.S. Department of Treasury, 2013). This would therefore increase the confidence in the financial market and increase lending. The Capital Purchase Program was intended to provide capital to financial institutions in exchange for preferred stock in case the SCAP showed a shortfall of capital. The extra capital would help financial institutions to absorb the losses from the mortgage defaults and the MBSs. With the extra capital banks would have a bigger buffer against losses and this should therefore provide credit to banks and households more easily, thus restoring the flow of credit in the U.S. (U.S.

Department of Treasury, 2013). Quantitative easing

In the period before December 2008 the Fed began lowering the interest rates in order to stimulate the economy as seen by the Taylor rates in Figure 4. This however did not have the desired effect of stimulating the economy (Mishkin, Matthews, & Guiliodori, 2013). With the official target rate at or close to zero, the Fed started to implement unconventional monetary policy in the form of Quantitative Easing (Federal Reserve Bank, 2014). In the period of 2008-2009 the Fed purchased long-term mortgage-backed securities, QE in this period is also known as QE1. In the period 2010-2011 the Federal Reserve conducted a second round of QE called QE2 and QE in the period 2012-2014 is called QE3 (Krishnamurthy & Vissing-Jorgensen, 2011).

QE1 had the primary goal of making it easier to obtain credit. By buying the MBSs the Fed increased the prices of the MBSs which lowers the yield on these assets due to the negative relationship

(21)

channel. In the period of QE2 the Fed mostly bought Treasuries, with the proceeds from the principal payments of QE1. The idea was that buying the government bonds would provide banks with liquidity. The increased liquidity should lower the risk premium asked by banks and therefore lower the long term interest rate to stimulate the economy.

QE can also affect inflation expectations. Since if the economy picks up there could still be a large money supply which could lead to a higher inflation. Thus raising inflation expectations and possibly increase current consumption and investment since the real interest rate would be lower.

4.3 Measures taken by the E.U.

4.3.1 Temporary measures

Enhanced Credit Support

The first immediate measures taken by the ECB was the Enhanced Credit Support (ECS). To give a better picture, at the end of 2008 the interest rate in the E.U. was cut from 4.25 percent to 1 percent. This however did not have the desired effect of restoring extension of credit and thereby stimulating the economy in the Eurozone, according to the ECB (ECB, 2009). Since conventional policy was not enough, the ECB implemented a set of unconventional measures to restore the flow of credit in the Eurozone. The set of unconventional measures were known as the Enhanced Credit Support. This section will focus on the temporary measures taken by the ECB to provide relief and to prevent the collapse of the

financial system (U.S. Department of Treasury, 2013). Therefore the following measures of the ECS will be discussed; the extension of the maturity of liquidity provision, fixed-rate full allotment, currency swap agreements and extension of collateral requirements.

In 2007 the European Central Bank (ECB) increased the maturity and the size of its Longer Term Refinancing Operations (LTROs) to provide euro banks with longer-term funds. When a bank is in need of funding it can ask the ECB for a loan, it should be noted that this is not the only way of financing. The loan must be backed up by collateral (European Central Bank, 2011). These loans are known as longer-term refinancing operations. Most of the LTROs mostly had a maturity of three and six months. The Extension of the maturity and provision program increased the maturity to 12 months. The goal was to reduce uncertainty in repaying debt and encourage banks to keep providing credit and to keep the money market interest rates low (European Central Bank, 2012). In December 2011 the 12 month LTRO was replaced by two longer term refinancing operations with a maturity of 3 years with the first

operation starting in December 2011 and the second in February 2012. The total expenditure was approximately 1000 billion euro (European Central Bank, 2011).

(22)

With the extension of the collateral requirements, the banks were allowed to use a wider variety of collateral to obtain liquidity from the LTROs. Just as their American counterpart, European banks had invested in ABSs. Since European banks were also exposed to the ABSs the goal of the extension of the collateral requirements was to provide banks with liquidity more easily. So that they could refinance illiquid ABSs and stop the liquidity shortages caused by the SIVs (Mankiw, 2010).

The ECB also adopted the fixed-rate full allotment which meant that banks had access to central bank liquidity at a fixed rate, the main refinancing rate, and against eligible collateral. Previously the ECB held auctions in which a fixed amount of financing was auctioned. The first bank who won the auction had to pay the highest interest rate and so on. With the fixed rate allotment the auction was obsolete and banks could get unlimited amount of financing (Mankiw, 2010).

A complementary way of providing liquidity was by currency swap agreements with the Fed. In a currency swap agreement the ECB provided liquidity in the form of U.S. Dollars. It provided funding in foreign currency against eligible collateral from the EU. This supported banks who would face a shortfall of foreign currency during the financial crisis (Mishkin, Matthews, & Guiliodori, 2013). Banks were facing a shortfall in foreign currency since the international interbank market dried up hence they could not get loans in foreign currency.

Another part of the Enhanced Credit Support was the Covered Bond Purchase Program (CBPP). Covered bonds are securities backed up by loans or mortgages. Prior to the crisis covered bonds were an important way of funding banks in the Eurozone. When the crisis unfolded the covered bond market dried up and banks were left with the illiquid covered bonds (ECB, 2011). Banks issued covered bonds to get funding. However, since the covered bond market dried up banks were now having difficulties in getting funding. This also led to a rise in spread in the covered bond market. With the CBPP the ECB bought the covered bonds to increase the demand for those bonds, leading to higher prices. As stated earlier higher prices imply lower yields. The lower yield lowered the spread between covered bonds and other bonds, thereby increasing demand. This would improve the liquidity of the covered bonds and would make it easier for banks to obtain funding since covered bonds were a way of getting funding (ECB, 2011). The Covered Bond Program was followed by the Securities Markets Program (SMP) in which there was an outright purchase of public and private bonds. The difference between QE and SMP was that the goal of SMP was to provide immediate relief to financial institutions whereas QE can be seen as

(23)

Bundesbank, 2011). With no change in the money supply there was less risk of inflation, which is the case with QE (Deutsche Bundesbank, 2011). The SMP was replaced in 2012 by the Outright Monetary Transactions (OMT). In the OMT Programme, the ECB could purchase sovereign bonds if the country would agree to implement structural reforms. Thus sending a message to the European countries that the ECB would act as a Lender of Last Resort, meaning that the ECB will still support countries and banks near bankruptcy. The aim was to bring bond yields down to lower the debt payments burden for

countries to ensure stability and positive expectations in the bond markets (Deutsche Bundesbank, 2012).

4.3.2 Structural measures

European Economic Recovery Plan

Structural measures are measures which try to restore the economy in the long run. The first measure to be discussed is the European Economic Recovery Plan (EERP), which was adopted in 2008. The goal of the EERP was to stimulate demand and increase consumer confidence. To do this an immediate budgetary impulse of €200 billion was needed (Commission of the European Communities, 2008).

Implementing a government stimulus package requires a significant amount of borrowing by the government. However, the Stability and Growth pact limits the potential investment in stimulating the economy by limiting the deficits of a country. It should be noted that the Stability and Growth Pact was agreed at E.U. level but the actual execution and implementation was at national level. To make sure that members of the E.U. kept stimulating the economy, the E.U. gave members the opportunity to deviate from the 3% debt to GDP ratio for two years. To deviate however members must implement structural reforms. Deviating from the 3% rule would give member states more room to focus more on stimulating their economies rather than focusing on the debt to GDP ratio (Commission of the European Communities, 2008).

In order to increase employment the EERP also included an employment initiative to reinforce activation schemes for low skilled workers and business start-ups, lower the social charges on lower incomes and would mostly concentrate on giving aid to the most vulnerable. The agreement also advised to reduce taxes on labor-intensive services. This would stimulate people to work and startup businesses and this would in turn increase their income, purchasing power and thus economic activity.

(24)

The E.U. advised to give incentives to invest in more eco-friendly technology by lowering taxes on green technology and promoted improvement of energy efficiency in buildings by reducing the property tax on energy-efficient buildings. The E.U. also advised to stimulate member states to promote R&D and education by providing fiscal incentives and grants. The goal was to keep stimulating

investment. The rise in investment would lead to a rise GDP (Commission of the European Communities, 2008). This concluded the chapter on the measures of the E.U. and U.S., in the next chapter the data and analysis will be discussed.

Chapter 5: Data and figures section

The purpose of thesis is to give an overall assessment of the measures taken by the U.S. and the E.U. after the unfolding of the crisis. In order to give an assessment in which economy performed better, several variables should be used like the gap with the NAIRU rate, real GDP growth rates and output gap. The NAIRU is the Non-Accelerating Inflation Rate of Unemployment and measures the lowest level of unemployment at which there is no increase in inflation (Mishkin, Matthews, & Guiliodori, 2013). The output gap measures the deviation of current output from its potential output. So it can be seen as an indicator for economic performance. The potential output is the output that can be generated if there was perfect competition and flexible prices (Justiniano & Primiceri, 2006).

5.1 Data source and description

The data on NAIRU, unemployment rates and output gaps are obtained from the databases of Organization for Economic Co-operation and Development (OECD). The data on real GDP growth rates and inflation rates for the Eurozone are from the database Eurostat, the database of the European Commission. It should be noted that in this paper the U.S. Is compared with Eurozone countries1)_._This makes the analysis more consistent since the data on the remaining countries was always available. The data on GDP of the E.U. is from the Eurostat website and the data on GDP of the U.S. is from the Fed.

5.2 Figures

As stated earlier the focus of this thesis is to provide an overall assessment of the measures taken by the U.S. and the E.U. after the unfolding of the crisis and to see if there is a performance difference between the economies. For this purpose the data on economic indicators like

(25)

5.2.1 NAIRU Rates

The analysis starts with a comparison of the NAIRU and unemployment rates. The NAIRU measures the minimum level of unemployment that does not cause inflation to rise. If the

unemployment rate is higher than the NAIRU for a couple of years the theory suggests that there will be deflationary pressures. The exact opposite holds for an unemployment rate lower than the NAIRU (Mishkin, Matthews, & Guiliodori, 2013).

Figure 5. The U.S. NAIRU and unemployment rates source: OECD

Figure 5 represents the unemployment rates and the NAIRU of the U.S. Figure 5 shows a rising yearly unemployment rate in the period between 2007 and 2009. In 2010 the spread between the NAIRU and the yearly unemployment rate was at its highest amounting to about 3%. In 2009 the U.S. government adopted the ARRA to decrease unemployment and stimulate economic activity. Since the ARRA there has been a reduction of the unemployment rates. Which could indicate the effectiveness of ARRA. The downward trend of unemployment continues and could indicate a movement towards the NAIRU rate. However, the decrease of the unemployment rate cannot only be attributed to the ARRA. In the same time period there were several other measures such as QE1 and TARP which both potentially have a stimulative effect on the economy or be counterproductive.

0,00% 2,00% 4,00% 6,00% 8,00% 10,00% 12,00% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

U.S. NAIRU and Unemployment rates

(26)

Figure 6 Euro-area NAIRU and unemployment. Source OECD.

As can be seen in Figure 6 the E.U. has followed the NAIRU with a maximum NAIRU-unemployment spread of approximately 1% up until 2011. The data does not show a decrease in unemployment, on the contrary the Euro area sees an increase in unemployment since 2008. A reason for the rise in unemployment could be the sovereign debt crisis.

So in terms of NAIRU and unemployment the U.S. has seen a a better recovery. However, this is only when looking at the data and it should be noted that the E.U. has seen a secondary shock and both economies have implemented different measures which also difffer in investment size. Therefore a direct comparison cannot be made.

5.2.2 Output Gap

The next reference value to be discussed is the output gap. The output gap measures the deviation of the actual output from the potential output. Which is calculated by deducting potential output from actual output. The potential output measures the potential GDP if all the economy’s resources were used (Gärtner, 2009). If the output gap is positive the aggregate demand is larger than the potential output which could create inflation. The opposite holds for negative output gap which could create deflation.

0,00% 2,00% 4,00% 6,00% 8,00% 10,00% 12,00% 14,00% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Euro area NAIRU and unemployment

(27)

Figure 7 Output Gap of the U.S. and E.U. Source OECD.

As can be seen in figure 7, the U.S. had a relatively large output gap prior to the crisis. This can be partially explained by the availability of credit before the crisis occur, since the availability of credit is positively related to consumption and investment (Brunnermeier, 2009). When the crisis unfolded the output gap decreased significantly to a negative output gap of -5%. The data shows that since 2009 there has been a trend upwards, this could be due to the ARRA which was implemented in 2009, QE and the TARP program. Even though the TARP program was implemented in 2008, most of the programs like the Credit Market Program (CMP) and the Bank Investment program (BIP) became active in 2009. The CMP and the BIP both tried to lower the interest rates for consumers and firms to stimulate demand. So the U.S. should be currently experiencing an increasing trend towards a positive output gap.

The E.U. has seen a slightly different trend, in the period of 2005-2007 the E.U. had a relatively high output gap with a peak higher than the U.S. in 2007. However, the drop in output gap after the crisis was less than in the U.S. Figure 7 also shows that there was an increase in output gap in the period of 2009-2011. Just as the U.S. the E.U. has seen an increase in inflation in the period of 2009-2011 and a complementary boost in output. However, the data also shows a drop of around 1.5% in the period of 2011-2012, most likely due to the sovereign debt crisis, and a minor increase of the output gap afterwards. Both economies are currently on approximately the same output gap. Since the

implementation of the measures there has been a steady trend towards a positive output gap in the U.S. whereas the E.U. still experienced a drop in output gap in 2011-2012. Again due to the second shock in the E.U., the sovereign debt crisis, a fair comparison cannot be made. However, it can be argued that the U.S. may have avoided a second crisis due to the recapitalization of the banking system.

-6,00% -4,00% -2,00% 0,00% 2,00% 4,00% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Output gap

(28)

5.2.3 Real GDP Growth rates

To continue the output analysis the real GDP growth rate will be discussed. The real GDP growth rate measures the growth rate of GDP while taking correcting for inflation. It therefore gives a more accurate look at the economic growth of a country since it only shows the actual GDP growth.

Figure 8 Real GDP growth rate of the U.S. and the Euro area. Source Eurostat and FED.

As shown in figure 8 the real GDP growth of the U.S. has been consistently higher previous to the crisis. After the unfolding of the crisis the real GDP growth of the U.S. dropped significantly just like the E.U. However, the U.S. real GDP growth rate decreased more than the E.U. in 2009. The data also shows that the real GDP growth rate of the U.S. increased to around its pre-crisis level, which could indicate the effectiveness of the post-crisis measures from the U.S. However, the data also shows a decreasing real GDP growth rate since 2010. This could be explained by the fact that a permanent increase in government spending has a negative impact on economic growth, even if the ZLB is binding (Cogan, Cwik, Taylor, & Wieland, 2009). The E.U. also sees an increase in real GDP after the

implementation of the measures. The decrease in real GDP means less production and therefore less demand for labor, figure 5 illustrates the rising unemployment rates in the U.S. in the period of 2007-2010. -8,00% -6,00% -4,00% -2,00% 0,00% 2,00% 4,00% 6,00% 8,00% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Real GDP growth

(29)

5.2.4 Inflation Rates

To assess whether the central banks kept their mandates the inflation rate will be discussed as well as the unemployment rate and the long term interest rates. The goals of the Fed were low

unemployment, stimulate demand and low inflation. The main goal of the ECB were price

Figure 9 inflation rates in the U.S. and E.U. Source Eurostat and FED.

Shown in figure 9 are the inflation rates of both economies. The data shows that throughout 2002-2007 the E.U. achieved their target inflation rate of 2 percent. When the crisis began there was a rise in inflation rate in both the U.S. and the E.U. Then a sharp decrease in inflation is noticed in the E.U. and deflation in the U.S. However, after 2009 there is again a trend towards the 2%.

Figure 9 also shows a rising inflation between 2009 and 2011 in both countries which could indicate a recovering economy since a higher inflation rate implies lower real interest rates and thus higher consumption. Hence higher production and demand for labor. Which can be seen by the decline in output gap in figure 7 and the increase in Real GDP growth in figure 8.

Just as the U.S. the E.U. shows an increase in inflation in the period from 2009-2011, an increase in output gap and an increase in real GDP growth. However, figure 6 doesn’t show a decrease in

unemployment. Figure 6 also shows an increase in unemployment from 2011-2013 which could be explained by the decrease in inflation causing the real interest rate to rise and thus making consumption and investment less attractive. This can also be seen by the decrease in output gap from 2011 until 2013 in figure 7 and decrease in real GDP growth in figure 8.

The unemployment rates have been discussed earlier with the use of figure 5 and 6. Figure 5 shows that since 2010 the U.S. unemployment rate has been decreasing and figure 6 shows that since 2011 the unemployment rate in the E.U. has been increasing.

-1,00% 0,00% 1,00% 2,00% 3,00% 4,00% 5,00% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Inflation rates U.S. and E.U.

(30)

To assess the goals of stimulating economic growth the real GDP growth from figure 8. As stated earlier currently the E.U. has a negative growth real GDP growth rate and the U.S. a positive real GDP growth rate. Thus in terms of Central Bank targets the U.S. has performed better in meeting its targets with a reasonably inflation rate of around 1%, decreasing unemployment and higher real GDP growth rates then the Eurozone.

5.2.5 Expenditure as percentage of GDP

To provide a better insight on the magnitude of the measures we will look at their expenditure as percentage of GDP. This will enable us to make a fair comparison of the measures.

Figure 10 shows that in terms of fiscal stimulus and QE (or similar) the U.S. has provided a bigger stimulation. The fiscal stimulus in the U.S. was about 4% higher than in the Eurozone. Also the increase in the Central Bank (CB) balance sheet has also seen an increase of 19% as opposed to the increase of 8% in the Eurozone. The expenditure on QE or similar measures has also been larger in the U.S.,

however it should be note that OMT is still active whereas QE 3 has been terminated. If the timing of the measures is considered it can be noticed that the ARRA and the EERP both have been implemented in 2009 so in this aspect there is not a significant difference.

Figure 10 expenditure as percentage of GDP. Source: Fed, ECB.

Type Fed Eurozone

Fiscal stimulus 6% 2%

Bank recapitalization 2% *

QE or Similar (total) QE1 12% CBPP 1%

QE2 4% SMP 2%

QE3 9.2% OMT -

CB balance sheet increase 2008-2014 19% 8%

(31)

Chapter 6: Discussion of empirical literature

To provide more insight on the effectiveness of the measures, the empirical results from the literature will be discussed.

Effectiveness of the fiscal stimuli

ARRA

Feyrer & Sacerdote (2011) investigated the overall effectiveness of the ARRA by examining the change in unemployment between February 2009 and October 2010. The effectiveness is examined at state level and at county level. The advantage of the state level analysis is that it provides more insight in the total effect of the ARRA on employment. It shows the change in employment per state when states are given a certain amount of stimulus. It only looks at the effect on the total output per state or total state employment, not considering the fact that there could be cross-state effects. County level analysis gives more insight in the state level analysis of employment by splitting each state into different counties. The county level analysis adds clarification to any fluctuation generated by the state level analysis. To calculate the effect on employment at state level Feyrer &Sacerdote (2011) use a cross sectional approach, with the employment rate as dependent variable and stimulus per capita as independent variables using the OLS method. The effects on employment at county level are estimated using a time series analysis.

After the regression at state and county level the data on employment rates are then further dissected to look at the effects of different types of stimulus spending on employment to see the difference in effectiveness of stimulus. For instance, did the stimulus have a greater impact on employment in education or construction? The downside of this approach is that spillover effects between states cannot be seen. This can therefore reduce the multipliers for some sectors.

The results at county level show an average fiscal multiplier of 1.06. This means that for every dollar that the U.S. spends for stimulation the output will eventually increase with 1.06 dollar .When the overall fiscal multiplier at county level is examined in more detail the results show that the expenditure on the low income households led to a fiscal multiplier of 1.9, the expenditure in education to a

multiplier of -3.31 and the investment in infrastructure multiplier of 1.85. The results at state level give an overall multiplier of 0.47. The impact at state level on the low income expenditure results in a multiplier of 2.31, in the Education sector the multiplier is -0.71 and the expenditure in infrastructure leads to a multiplier of 1.85. If the results at state level and county level are compared, it can be seen that the sizes of the multipliers differ quite a lot especially the overall multipliers. A fiscal multiplier

(32)

below unity would mean that the expenditure would cost more than it would benefit the economy. The results also show that the effectiveness depends highly on the type of spending. The results show positive multipliers in most sectors except the education sector, which has a highly negative multiplier which means that it negatively affects output. A possible explanation for the negative education multipliers is given by Feyrer & Sacerdote (2011). They state that the State Departments of Education advised schools to avoid making permanent hires since the grants were only temporary. Thus schools used the money to keep the current staff rather than hiring more staff and set the rest aside as backup which has a negative effect in stimulating the economy since it decreases consumption since

unemployment stays relatively equal.

A drawback of this empirical result is that the estimated effects are limited to the first three quarters after the crisis and that the effect of the ARRA could possibly be noticed even after the sample period, meaning that this may not be the full result of the ARRA.

Chodorow-Reich, Feiveson, Liscow and Woolston (2012) investigated the effect of the Medicaid expenditure of the ARRA on unemployment. Medicaid is a social healthcare program for low income households. The ARRA raised the overall budgets for states by providing extra funding for Medicaid, this should avoid additional budget cuts and thus provide fiscal relief for states. Thereby preventing extra cuts in state spending that would lead to demand fallout and thus lower output. Chodorow-Reich, Feiveson, Liscow and Woolston (2012) estimated the effect of the Medicaid expenditure on employment using a cross sectional analysis at state level. This was done using OLS and IV regression. The data set contains monthly employment data from June 2000 to June 2010. By using a cross sectional analysis however, they also ignore the spillover effects in different states, just as Feyrer& Sacerdote (2011). For example, a rise in spending in state A may lead to a higher demand in state B and thus raising

employment and production in state B. This then means that the results might underestimate actual nationwide effect of the ARRA on employment. They find that per $100.000 ARRA expenditure employment increased by 3.8 job-years this corresponds with an overall fiscal multiplier of 1.5. Cogan, Cwik, Taylor & Wieland (2012) used the Smets-Wouters macroeconomic model, calibrated to the U.S. economy, to assess the effects of the ARRA. The Smets-Wouters model can be seen as a New-Keynesian model, which assumes that consumers and firms have rational expectations and that there is a form of price rigidity. This captures how people’s expectations and behavior change

The effectiveness of the post-crisis measures in the United States and the Euro-zone