| Profil von SeamanMy worldFotosBlogListen |  Extras  Hilfe |
My world31.12.2008 新年快乐回眸百年历史，形势一片大好。
1919 China May Fourth Movement: People ask “Where is China?” 1929 The Great Depression: People ask “Where is Capitalism?” 1949 Chinese Revolution: Only Socialism can save China. 1979 Deng Xiaoping’s Reforms: Only Capitalism can save China 1989 Fall of the Berlin Wall: Only China can save Socialism 2009 Global Financial Crisis: Only China can save Capitalism. 2019 ??: ???
Can China save capitalism? Yes, it can. In October, China (mainland) holds $652.9 billion US Treasury securities, surpassing Japan to be the largest US Treasury debt holder. The U.S. may have to issue as much as $2 trillion in debt over the next four quarters to pay for the $700 billion financial-system bailout, the war efforts in Iraq and Afghanistan and Obama administration’s economic stimulus package. As the global financial crisis is worsening, the US Treasurys, despite of their low yields, seems the safest investment for China to prevent the appreciation of the yuan, given its $2 trillion foreign exchange reserve, the largest around the world. So China may save capitalism indirectly.
In another point of view, China can save capitalism by setting an example for “big government”. Before the crisis, the Chinese bureaucratic control over its financial industry and public-owned economy is often being criticized for its lack of free-market spirit. Now the westerns are paying for their privatization and deregulation as the housing bubble and credit bubble burst. The role of “big government” is set to become larger in the US and Europe through the massive aid from the public sectors, for example, the nationalization of Fannie Mae and Freddie Mac, and twice bailouts of AIG. Although many economists claimed that the government interventions with fiscal stimulus and tax cuts follow John Maynard Keynes’s idea, the methods used by both China and the west coincides to be the same at least for the current situation.
However, China has its own problems. In November, the Chinese exports, which have contributed about 25% to GDP over the last couple of years, dropped 21.4% compared with October. It is the first monthly drop in seven years. Chinese crude steel output in November also fell 12.4% from a year earlier, pushing the consolidation within its steel industry to shield from financial impacts. With the deterioration in consumer spending in US and Europe, Chinese export will suffer greater losses. Also, deflation is looming as CPI declined from 8.7% to 2.4%. Morgan Stanley CPI forecast for 2009 will down to -0.8%. To avoid an entrenched deflation expectation, a decisive monetary policy is needed. The key deposit interest rate will drop to below 1% next year according to JP Morgan. The truth is that China is far from being decoupled from the global financial crisis; its export-dependent economy is vulnerable.
The good news is that no one can quickly and easily implement such a big economic expansion like Chinese. Recently, Chinese policy-makers response the financial shocks quickly by (1) easing monetary policy by cutting deposit interest rate to 2.25%, (2) introducing an aggressive economic stimulus package of RMB 4 trillion (US$586 billion) to boost domestic demand, (3) keeping the exchange rate low (6.85 RMB/USD) to support exports, all in an effort to maintain GDP growth of around 8%, which is said to be the minimum speed to absorb all the new workers coming on to the market. Morgan Stanley’s forecast for China GDP growth in 2009 is 7.5% given that the expansionary policy takes into effect and real estate industry does not collapse.
The large amount of investments from the stimulus package boosts infrastructure constructions, real estate industry, steel production, transportation, etc, and is the most efficient way to lift the economy up. Figure1 shows that the share of total investments to GDP jumped from 22% in 2000 to more than 50% in 2001 since the US recession drove the capital flowing into emerging markets including China. The amount of total investments keeps increasing since then while the total consumption increases mildly, whose shares to GDP decreases over the periods. Also, net export began to help economic growth in 2005 as the US and Europe are enjoying cheap credit, pushing the demands higher. We shall expect total investments in 2009 to play a more important role in economic growth as both net exports and domestic consumption shrink.
 Figure 2
 But the effects of the stimulus package will not show up until the second half of next year. Also, as the stimulus package will boost infrastructure investments, its main beneficiaries are state-owned enterprises. Private sectors will suffer a tightening credit from banks as well as big drops on export orders. The uncertainty in real estate market is increasing, putting the pressure on economy recovery so that local governments, like Shanghai, are busy in enacting attractive rules to stabilize the real estate markets. The overall economic outlook next year is expected as “getting worse before getting better”.
Perhaps, a recession can be averted technically in China, but its less resilient political system is going to be tested before the economy recovers not any time soon. One journalist said recently in Financial Times that “it would be a historic irony if the Chinese Communist party was thrown into crisis, not by the collapse of communism in 1989 but by the convulsions of capitalism in 2009.” An economic problem can become a big threat to social stability in China. To avoid chaos, the current stimulus package is necessary. But in a longer term, we shall expect China’s economy to shift from export-dependent to domestic-consumption-driven. Such economic transformation can reduce the cyclic shocks from abroad, and keep growth sustainable. Last Friday (Dec. 26, 2008) at 2008 China Financial Forum, Zhou Xiaochuan, the governor of People’s Bank of China, also urged to increase domestic consumption demand to minimize the impact of the financial crisis. He agreed that the share of consumption in GDP now is too low (about 50%) and consumption structure is unbalanced. And he suggested that a broad and complete consumption-supportive policy is needed to boost consumption, in which many other factors like individual consumption mood, consumption expectation, social welfare, and so on.
Although these factors are necessary to the economic transformation in short term, or for next 10 years, the strengthening of individual consumption in long term, which is the key to the domestic-consumption-driven economy, cannot be limited to these factors but requires overall reforms including financial industry, social structures and politics.
1. A more efficient banking system, or broadly speaking, a modern financial service industry, as a financial intermediary can enhance finance channels for both corporations and individuals, and thus boost consumption. Restructuring the banking system will force the state-owned enterprises to reform as well through the credit and finance channels and tax policy.
2. Urbanization, the most fundamental factor for China’s economic development, is responsible for high GDP growth rate. But it also widens the gap between the urban and rural areas. The process of urbanization should be linked to raising standard of living of people, especially those who live in rural areas. They are the majority of population and will have the biggest market potential and purchasing momentum as the economy grows.
3. The state-owned social welfare system including healthcare, unemployment insurance and retirement has to be reformed so that a large share of income will go to consumption instead of saving for future uncertainty.
4. The above are just selected aspects, but all require an advanced legal system. For example, the household registration system, or hukou, is a barrier in the economic transformation, and should be reformed. Also, urbanization needs new regulations in land utility, etc. Furthermore, an improved social welfare system aims to allocate welfare more efficiently and thus boost consumption. This is not limited to economics issues but is an important part of national development strategy related with political system.
Over the last decade, China’s urbanization moves forward rapidly. However, as the urban population increases, the inflation-adjusted gap of wealth between the urban and rural areas widens (see Figure3). The black line shows the increasing gap, and has a kink in year 2001, which is consistent with the jump of total investments in Figure1. Most of the investments are poured into urban areas, widening the wealth gap. Figure4 further supports this point. For example, the annual percentage change of total investments reached its peak in 2001, and the wealth gap has its biggest positive change a year later. In other words, the overall investment, which is the major player in the economic growth, does not benefit rural areas. Also, with a strong export in 2007, the share of total investment dropped. We would expect a deceleration in the widening wealth gap between urban and rural areas in 2008. As Prasad and Rajan (2006) point out that China has reached a stage of development at which many key reforms are unavoidably interconnected. The current stimulus package is an emergency alternative, which cannot change the situation in longer term. As the government’s balance sheet is expanding rapidly, the battle with inflation then will be even tougher since inflation will come back as quickly as it left three months ago. With the scale for post-crisis economic development to be more enormous over next decades in China, there is less room for the government to make mistakes. Figure3 Figure4
Reference
Prasad, S. Eswar, and Raghuram G. Rajan, March 2006, “Modernizing China’s Growth Paradigm”, forthcoming in the American Economic Review. Rachman, Gideon, December 16, 2008, “China's economy hits the wall”, Financial Times. 24.11.2008 Administration's New Chief Financial SpokesmanThe US President-elect Barack Obama's selection of Timothy Geithner as his Treasury Secretary today boosted the stock market again. The total jump in DJIA since last Friday is more than 800 points. Obviously, investors are satisfied with Mr. Obama's choice, and they expect the US Treasury led by the new secretary together with the Fed can steer the US economy to overcome the current turmoil. Mr Geithner is not an economist but he leads a team of economists at New York Fed to tackle the hardest financial problems people have ever had, including the emergent rescue of Bear Stearns in March, fall of Lehman Brothers, merge of Merrill Lynch with BOA, rescue of AIG in September and so on. He is a very sharp guy in terms of practice and sensibility. Every time in the weekly meeting at New York Fed, I saw him reply to economists with a number of questions from various points of views on one issue after economists reported the market conditions. From the talks, I can feel that he is highly pragmatical, and looks into problems beyond any economic theories although he knows enough economics to understand the basic theories involved. Now, the first big task he is going to face is the use of remaining $410 billion TARP funds. The worst scenario will be that the economy does not have any improvement when the funds are used up. The whole market is waiting for the next administration's actions. As Mr. Obama said in his economic team announcement today that "These extraordinary stresses on our financial system require extraordinary policy responses. And my administration will honor the public commitments made by the current administration to address this crisis." Congratulations on the nomination and good luck, Tim!
13.10.2008 Krugman's International TradePaul Krugman
John Nash joined in the celebration of Krugman's Nobel Prize in Economics at Woodrow Wilson School of Princeton University.
14.09.2008 中秋节快乐!Tonight will be a remarkable night. When I was celebrating the Mid-Autumn Festival with my friends, something was happening at the New York Fed in lower Manhattan. Now the dust is down; Bank of America is going to buy Merrill Lynch for about $29 a share. Lehman was working on a possible bankruptcy filing and has to liquidate its assets then. But the future of about 25,000 employees at Lehman and an additional 60,000 at Merrill is up in the air, and the liquidation will cause another huge downward spiral over the markets. The downward spiral in asset prices will not stop until most of companies' over-evaluated assets have been written down and their balance sheets are rebuilt. So, the only good news is that the quicker the market hits the bottom, the sooner it can bounce up hopefully early next year. But who will be the next to collapse before then?
16.08.2008 写在离开美联储的日子8月8日的开幕式的确很精彩，张艺谋没有给国人丢脸。唯一美中不足的就是假唱事件。孩子是无辜的，说到底还是大人们的思维观念问题。看看网上的议论就明白，公道自在人心。大地震之后，我们看到了久违的‘以人为本’的精神。现在中国敞开大门，欢迎五湖四海的朋友们来作客，那就应该把门开得更大些，这也正是奥林匹克精神所追求的,不是吗？
8月15日是我离开美联储的日子。从下个月开始，我的出没范围将向西南方向移40英里，到一个叫Princeton的小镇上去。在那里有个“泡泡实验室”,将是我后两年学习，生活的地方（http://www.princeton.edu/bcf/newsevents/in-the-news/links/20080516-Bernankes-Bubble-Lab-WSJ.pdf）。 两年的工作让我懂得什么是research。research = re + search = search again and again。更重要的是，从去年开始的金融危机中，让我见识了处在风暴中心的央行是如何应对这一世纪灾难的。所以，我把我两年来看到的，听到的，以及想到的纪录下来，只是为了整理一下自己的思路，以免日后忘记。当然，如果您看了之后，有什么想法也可以告诉我。
The following is my own view about research and the markets over last two years. It is preliminary, and comments are welcomed.
Working in research department at New York Fed is my first job in finance. And my first task, which finally turned out to be my main task for two whole years, was to develop term structure models on interest rates. The purpose of term structure models is to understand what moves bond yields across maturities, or how changes at short end can transfer to long-term rates.
As we know that in every Federal Open Market Committee (FOMC) meeting, the Fed sets a target interest rate, or to be precise, the federal funds rate that financial institutions charge each other for overnight loans. In the meeting of August 5, 2008, which is the last FOMC meeting before I leave the bank, the target rate was unchanged at 2%. I will talk about its meaning later but simply put, the Fed can only effectively control short-term rates. However in macroeconomics, aggregate demand is driven by long-term rates. For example, people make decisions on purchasing houses or cars based on long-term mortgage rates instead of overnight rates.
According to the pure rational expectation hypothesis, long-term rates should reflect market expectation for the average future short-term rates, or in economic terminology, the forward rates should be equivalent to the future expected spot rates. But this is not always true. A more general hypothesis assumes that the difference between forward rates and expected spot rates is the term premium, which is required by investors for the compensation of investing in longer maturities. Unfortunately, both expected spot rates and term premium are unobservable. So, term structure models can help us to decompose interest rates into expected spot rates and term premium across maturities.
Based on these features, term structure models can be naturally applied to any interest rates related financial derivatives, like options, credit default swaps (CDSs), collateralized debt obligations (CDOs), and even currency exchange rates (Pan and Singleton 2006, Liu, Longstaff and Mandell 2000). Nonetheless, we should be cautious that any mathematical models including term structure models can only explain few aspects of the whole market. As Fed chairman Bernanke said:" No matter how sophisticated the theory behind an estimated equation, the equation is always naive. This is particularly problematical in the social sciences, when the basic unit of observation is so often that political, free-willed, unpredictable character, the individual human."
From the subprime crisis since last summer, people realized that the development of financial derivatives outpaced their understanding. All participants in the market are tightly woven together but they do not know clearly where the risk comes from and how big it is. People do not know what they do not know, otherwise, the crisis would have been avoided. I will go back to this point later, but let me further the discussion on the term structure models with applications in monetary policy.
In 2005, then Fed chairman Alan Greenspan provided his view about term premiums in yields when responding to Jim Saxton, chairman of Joint Economic Committee (JEC) (Greenspan 2005). He said:"...The difference between near-term and long-term risk premiums ... is far less certain and likely to depend on economic circumstances." Then, his successor, Ben Bernanke followed a similar tone by saying in 2006 that "I would not interpret the currently very flat yield curve as indicating a significant economic slowdown to come, for several reasons. ... to the extent that the flattening or inversion of the yield curve is the result of a smaller term premium, the implications for future economic activity are positive rather than negative" (Bernanke 2006). Both of them doubted the forecasting power of term premium.
But based on Estrella and Wu's three-factor term structure model (Estrella and Wu, 2008), we find that term premiums contain useful information about future real economic activity not included in pure expectation although the decomposition of nominal yields into expectations and term premiums may not significantly enhance predictive power of, for example, forecasting recession. And, I want to introduce Arturo Estrella's recession forecasting model, which opens my eyes about what a good model should be.
It is a probit model by regressing nominal 10-year and 3-month yield spread on recession indicators 12-month ahead, and then calculate the probability of recessions (see Estrella and Hardouvelis 1991, Estrella and Mishkin 1996, Estrella and Trubin 2006 for details). From 1960 to 2005, there are total six recessions in US history defined by National Bureau of Economic Research (NBER), all of which were caused by different events. But the model can forecast the probability of recession over a long period (12-month ahead) using just nominal yield spread without considering the cause of the recession!
Figure 1 shows that the probability of recession surged above 30% in each recession (shadow area), and at the beginning of 2007, the model signaled a likely recession in 2008. Although NBER has not announced recession yet, most of the major economic indicators show that we may be already in the recession or on the edge at best. The figure also shows that the probability of recession goes to near zero in 2009, so we will all be safe next year.
From my experience in two years, I think a good model should be 1) practical. People use simple and explicit mathematical models to mimic the dynamics of the market. As Wall Street saying goes:" if it may be true in theory but it won't work in practice, get a better theory." In the subprime crisis, for example, risk models failed to evaluate complex securities, and forecast potential losses. A likely reason is that the data on underlying collateral is very scarce, so new models should fix this problem. 2), a model should be as simple as possible but never simpler. There always exists a trade-off between simplicity and completeness. Mathematical models work in a low-dimensional space but to explain a complex real world. So, the trade-off depends on which aspect the model works with, or what assumptions the model has, i.e. what must be modeled and what can be approximated. The more factors a model needs to calibrate to, the less useful the model is. 3), a good model should also be innovative. It can be new in algorithmic techniques, or it can conceptually change people's view.
Based on these three criteria, there are several candidates of good models like Black-Scholes model, Taylor rule, etc. And of course, Arturo's recession forecasting model is definitely one of them.
Black-Scholes model, for example, which has only few inputs like strike price, risk-free rate, etc., is also a good model. It is now widely used in the market. More important, the concept of implied volatility obtained from the model opens my eyes. An analogy between this and temperature explains the reason clearly. Since temperature is invisible, people can know it only through a device like thermometer. Similarly, volatility is unobservable, and we know that actual option prices are rarely consistent with the ones from the model due to its idealized assumptions. So, we can use the model to find implied volatilities given option prices from the market. By doing so, we are able to 'see' the volatility and further trade on that. The implied volatility from S&P500 index options, or VIX, is a popular measure of equity market volatility. Figure 2 shows the history of VIX since 2004, which catches all financial shocks in the market.
Another lesson I learned from the work is that the expectation plays a significant role in the market. Many economists, like Robert Lucas, Christopher Sims, etc, are trying to model people's expectations in macroeconomics. Since participants in the market make decisions based on their expectations in the future, the turmoil in the market, for example, high food and oil price, and the fire-sale of Bear Stearns, can all be explained more or less by the expectations but in different forms. For example, people usually think that expectation on inflation can be implied by the gap between nominal and real yields, or Fisher equation, as:
Implied inflation = nominal yields - real yields
Before the issuance of CPI index-linked bonds, or Treasury Inflation Protected Securities (TIPS), real yields were unobservable. But now, TIPS can be a good proxy. The logic behind implied inflation is that if investors worry too much about inflation, they will buy inflation-protected bonds to keep their money safe. When more people make such flight of quality, the spread between nominal and index-linked bond yields, or implied inflation will increase, thus reflecting investors' expectation. However, implied inflation will often be overestimated due to a time-varying inflation risk premium, a compensation for investors to bear the inflation risk. So, generally, the following equation holds:
Implied inflation = nominal yields - real yields = expected inflation + inflation risk premium
In the paper of Adrian and Wu (2008), we developed a term structure model on both nominal and TIPS yields in order to break down implied inflation into expected inflation and inflation risk premium. In order to get a convincing inflation risk premium, which is highly correlated with the covariance between inflation and real factors, our model not only fits the level of yields but also their second moments.
To find a reliable inflation risk premium is important, because, first, since the average inflation rate is largely determined by monetary policy, the variation of inflation risk premium in long-term bonds is closely tied to the market's expectation on the policy. From a rising risk premium, policy makers can be alerted to act promptly and decisively. Second, for investors, they may think about how to protect, or even make profit from the rising inflation risk. We find that the forward five to ten year expected inflation, a better measure for forecasting inflation than implied inflation, anchored at 2.3% as of July, while the inflation risk premium decreases to 0.5% from its highest of 1.2% in March 2008. (see figure 3).
Admittedly, the shortcoming of this model is that the pool of index-linked bonds is smaller and less liquid than that of nominal bonds, even though we mainly focus on last five years of TIPS data. In addition, CPI itself, with which TIPS is indexed, is suspect as Wall Street saying goes: "inflation is everywhere except in the CPI." Nonetheless, the results from the model are interesting. For example, the forward five to ten year inflation risk premium is highly correlated with volatility index in bond (MOVE) market (see figure 4).
In a word, the implied volatility, the implied inflation, and state variables in the term structure models together with some very useful tools, like Kalman filter, GARCH, and maximum likelihood estimation, open my eyes in research. The purpose of using them is to find invisible factors that are under the chaotic market activities but drive the market dynamics. However, no one has perfect knowledge about the whole market. The implied volatility, for example, is just one of many ways to show how volatile the market is.
Moreover, these unobservable factors can affect one another so that it is very difficult to extract any one of them alone. Market is complex, especially during the crisis. These days, both economists in the academics and practioners in the industry are trying their best to explain what happened since last summer. And meanwhile, they are forecasting what will happen next, which is even harder. As former Fed chairman Alan Greenspan said:"... forecasters' concerns should be not whether human response is rational or irrational, only that it is observable and systematic." But how to observe the unobservable factors or mechanism? That lies in the center of the research of the crisis.
Figure 1
Figure 2
Figure 3
Figure 4
The subprime crisis, in my point of view, exists four unobservable factors that distort the market activities, including liquidity risk, credit risk, counterparty risk, and a downward spiral of asset price. All of these factors exist together, and affect one another. This kind of co-existence of risks can quickly spread shocks of one company to overall market, and then across markets and countries. However, these risks are not from nowhere. They hide in the dark side of financial innovations and can be amplified by people's expectations, pushing the market even further from normality. The financial innovations created the credit boom, which, together with the housing boom, were two sources of the financial turmoil according to Fed chairman Ben Bernanke (Bernanke 2008). The two booms were enforcing each other before bursting.
The U.S. housing boom started in mid-1990s and reached its peak in 2005, and the house prices stopped accelerating around the same time (see figure 5). Based on the mortgages, structured products were created via securitizations. Simply put, securitization allows people, usually banks, to collect pools of mortgages, loans, and corporate bonds with different level of risks, or trenches, which can be sold separately to different investor groups. Collateralized debt obligations (CDOs) are the most common form of structured products. Others include collateralized loan obligations (CLOs), and collateralized mortgage obligations (CMOs). Securitization became popular among investors, banks and brokers because the long-term illiquid assets that banks have can be sold through short-term securities, and these securities with various trenches can meet different investors' needs. Also, by purchasing high rating trenches, banks can have some benefits, i.e. to improve their balance sheets or do rating arbitrage (Brunnermeier 2008). Thus, structured products expanded on a dramatic scale.
Furthermore, the market is too reliant on credit ratings. Rating agencies' job is to evaluate pools of mortgages and give ratings so that banks can sell securities backed by different quality of mortgages to different investors. A rating can make or break a deal; a downgrade can trigger a bankruptcy. But they judge the quality of mortgages mainly based on historical patterns of default. And when market becomes chaotic, the statistical models cannot work properly because they live in a world where historical data cannot provide any forecasting information. As Wall Street saying goes: "(the model works as to) use weather in Antarctica to forecast conditions in Hawaii" (Lowenstein 2008). By doing so, rating agencies lulled people into a false sense of security by giving top ratings.
The proliferation of securitization is also due to a severe agency problem according to Frederic Mishkin (2008), in which the agent, or the originator of the loans, did not have the incentives to act fully in the interest of the principal, or the holder of loans. Banks thought that securitization helped them shift the risks on the loans and mortgage to investors. But the fact is that credit risks were still in the banking system, causing a huge loss for banks during the crisis. Nonetheless, when times are good, such originate-to-distribute business model encouraged banks to take more loans by loosing credit standards to unqualified home owners, or subprime borrowers. Since they cannot afford higher rates, they would need to finance soon. And to earn substantial fees, banks are willing to help borrowers based on the expectation that their house prices will increase.
This is a classical origin of bubble studied by "Bernanke's bubble laboratory". The "laboratory" consists of a group of professors at Bendheim Center for Finance at Princeton University described by Wall Street Journal (Lahart 2008). They are doing researches on financial bubbles under the leadership of Ben Bernanke, then chairman of the center. And they concluded that this self-enforcing cycle in expectation discouraged most players, including banks, brokers, or investors, to bet against the rising trend in house prices because to go short on prices is very risky, and it is easy to make money by just following the trend at least before the bubble blows up. When most skeptics stay out of the market, and only optimists are playing, the prices go higher and higher. As CitiGroup's former CEO Charles Prince said:" When the music stops, in terms of liquidity, things will be complicated. But as long as the music is playing, you've got to get up and dance". A bubble thus comes.
Another potential problem is that banks relied heavily on short-term funding while investing in illiquid long-maturity assets. Generally, banks finance their balance sheets through four markets according to Brunnermeier (2008). First is asset-backed commercial paper market, in which commercial paper is backed by the underlying assets, allowing the owner to sell the collateral assets in case of default. Second is inter-bank market, in which banks borrow and lend to each other at a certain rate. The average of the rate is Eurodollar LIBOR rate, regarded as a good measure of cost of borrowing between banks. The third is fed funds market, in which banks lend the excess reserves to others to earn interest. The last is repo market. In the repurchase agreement, or "repo", banks can get funding by selling their securities to the Fed or other banks and agree to repurchase them back later. Usually, banks use their credibility to ensure the availability of these funding. If the credibility is losing, banks will be in trouble of funding.
When houses prices were too high to be affordable, the expectation of rising house prices faded. Subprime borrowers were the first group to suffer the losses, and were then forced to foreclose. The housing bubble was bursting. Figure 5 shows that since 2006, both the number of house sold and change of house prices plummeted. Both of them reached their lowest level recently, which were down about 63% and 15% from their peaks respectively. The weak housing market triggered a sharp slowdown of issuance of subprime loans, and that affected the structure products in the credit markets whose underlying were these loans. On August 9, 2007, BNP Paribas was forced to freeze three investment funds because of mortgage losses. Also, Countrywide Financial, one of the biggest mortgage loan brokers announced its funding problem, which increased the uncertainty about true value of these products, and triggered a liquidity crisis. Although the crisis originated in subprime mortgage, it had a big impact on other parts of the mortgage market.
In US, there are four categories of mortgages: the major part is prime mortgage conforming to underwriting standards set by Government Sponsored Agencies (GSEs), like Freddie Fac and Fannie Mae. Others including non-conforming (Jumbo), subprime which involves with borrowers with poor credit history, and Alternative-A mortgage whose borrowers are between prime and subprime. In the boom, subprime mortgages increased rapidly. From 2001 to 2006, the origination of subprime mortgage was more than doubled to 600 billions while prime mortgage decreased 40% to 1040 billions (Ashcraft and Schuermann 2008). As the uncertainty increased, investors lost confidence on subprime mortgage; they also began to suspect the valuations about other mortgages. Even worse, in September 2007, rating agencies began to modify their methodologies for these securities, further exacerbating investors' expectations. As a result, the mortgage market quickly dried up. For example, as one of the major sources of funding for financial institutions, the issuance of asset-backed commercial paper sharply declined in July 2007 (see figure 6). And until recently (August 2008), the aftermath of the crisis spread to prime mortgages so that the US Treasury has to launch a rescue plan for Freddie Fac and Fannie Mae.
In interbank market, the conditions also deteriorated. On the day of Paribas's announcement (August 9, 2007), the spread between 3-month Libor and forward 3-month Overnight Indexed Swap rate (OIS), or the LIBOR-OIS spread, suddenly jumped from 6.4 to 55.4 basis points (see figure 7). The LIBOR-OIS spread reflects the availability of funds in the market. So, the jump in the spread signaled that traders are worrying about banks' ability to refinance their balance sheets. And credit default swap (CDS) spreads for these companies rose sharply, implying an increasing concern about their risks of default. The spread for Bear Stearns during its fire-sale, for instance, rose to 7.37%, the highest level among others, meaning that annual cost to insure 10 million dollars of its debt would be 737,000 dollars (see figure 8). The spreads have improved since the Fed's actions but still remain elevated compared with a few months ago.
With the concern about counterparty risk and liquidity risk, banks were unwilling to lend funding to each other. Thus the Fed launched Term Auction Facility (TAF) on December 12, 2007, which combined features of open market operations and the discount window's primary credit loans in an effort to provide more liquidity into banks (Armantier, Krieger, and McAndrews 2008). However, the market deteriorated further in January 2008 due to the aggressive unwinding of Societe General's 49.9 billion Euro position triggered by its rogue trader.
Until then, banks had faced pressure of funding from several sources. First, risk-aversed investors pulled back their capital from their counterparties in anticipation of the potential withdrawals of funds by their own investors. In other words, the expectation about risks increased so that everyone wanted to pull back money before everyone else. They knew that if they lagged behind, they would get much less than what they would get now. This kind of high-order expectation caused the liquidity in credit market to disappear in a second. Also, the turmoil in commercial paper market and inter-bank market shut down a large part of funding sources. More important, de-leveraging amplified the impact. When times are good, the four funding markets are very liquid, and the cost of borrowing is low. So, banks borrow more money to multiply their profits. According to Adrian and Shin (2008), financial institutions increase leverage in booms and reduce in bust because they buy more when prices are high in the expectation that price will go even higher, and sell when prices are low in the opposite expectation, which, thus, exacerbate the market fluctuations.
Figure 9 shows that during the crisis, all top five investment banks de-leveraged their balance sheets after reaching their highest leverage ratios. At Bear Stearns, for example, the ratio remains high when comparing with others since 2001, and became the highest right before the collapse. Merrill Lynch started to sharply leverage since the third quarter of 2006, and increased more than 50% to 33 of its peak, which means that total assets were 33 times the value of shareholders' equity. On the other hand, if required, it has to contract its lending by a multiple of its credit losses in order to restore its balance sheet. Up to July 2008, it has written down about 52 billions of assets, only next to Citi Group (see figure 10). Both of the two financial institutions owned a much bigger amount of CDOs than their peers.
Actually, in March, the fire-sale of Bear Stearns to JP Morgan in March 2008 gave a vivid picture of how these problems above triggered a storm that buried the fifth largest investment bank of US in one week. Its story (Bryan Burrough 2008), in my point of view, is one of the most exciting financial events just next to the fall of Long Term Capital Management in 1998 (Lowenstein 2000).
Two hedge funds at Bear Stearns Asset Management were first run into trouble. By leveraging astonishing 100 times their available money to trade, two funds suffered a huge loss on MBSs as housing market deteriorated in 2007. This led to $1.9 billion write-down in 3rd quarter of 2007, which directly drove its CEO, Jimmy Cayne, out. However, this did not help Bear Stearns. Since February 27, 2008, the rumor about Bear's financial health widely spread, and severely shake investors' confidence. On the next Tuesday, March 11, many of Bear's major counterparties, including Goldman Sachs, Credit Suisse, and some big hedge funds, stopped trading with it. As a result, Bear realized that there would be $30 billion loans that would not be rolled over the next morning. They could either find emergency cash infusion or make a bankruptcy filing. By then, its share price has plunged 66% from $87 a week before to $30 per share then (see figure 11). Finally, Bear turned to Fed for help. By Sunday night, Bear Stearns had struck a deal to be sold to JP Morgan for $2 a share, and the Fed had agreed to lend Bear $30 billion, backed by the assets on its balance sheet. And the Fed announced a new credit facility to primary dealers in order to avoid Bear-Stearns-like case to happen again.
The rescue of Bear Stearns was regarded as a critical point in the subprime crisis since it decreased the counterparty risk and maintained the function of the market. Like investment banks which expand in boom and contract in bust, hedge funds acted even more aggressively. In September 2006, New York Fed President Tim Geithner already raised his concern about the role of hedge funds in financial markets (Geithner 2006). He foresaw the possible large damages caused and amplified by individual hedge fund actions. The actions had the aggregate effect of inducing even larger price declines and surge of risks, leading to systemic crisis. As a matter of fact, hedge funds are not culprits in this crisis although they amplify the financial shocks passively. Many funds, including the two of Bear Stearns, were forced to de-leverage and unwind sizable positions when facing high haircuts and margin calls, causing a downward pressure on overall markets.
Generally speaking, hedge fund industry is resilient during this crisis. Average hedge fund performance since last summer has not fallen significantly, but the dispersion of returns across and within strategies is remaining high recently after reaching to its highest level in December 2007 (see figure 12). Strategies of global macro, emerging markets, and short bias have the best performance since last year, which return 26%, 15%, and 16% respectively, while fixed income and convertible arbitrage strategies do not perform well according to TASS Hedge Fund Index by strategies (see figure 13). The big loss in fixed income arbitrage strategy which usually profits from the convergence of price to its fair value, for example, was mainly due to the high volatility of the turmoil in the markets. And in March 2008, industry-wide returns dropped significantly because of another tide of de-leverage and assets write-down.
However, individually speaking, the richest hedge fund managers keep getting richer. Someone's pain is someone else's gain. John Paulson who made $3.3 billion last year, for example, stays at the top among richest fund managers (see figure 10). As early as 2005, he began betting that complex mortgage investments would decline in value. Someone who shares his view includes George Soros with his world-famous Quantum Fund, Philip Falcone of Harbinger Capital funds and Stony Brook alumni James Simons of Renaissance Technologies, all of whom earned $2.4, $1.7 and $1.3 billion last year by shorting subprime credit. Meanwhile, some managers like T.Boone Pickens and John Arnold, made their fortunes by betting on energy, which returned $1.2, and $0.7 billion respectively.
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
(see Aug 1 blog)
Figure 11
Figure 12
Figure 13
Expectation plays a vital role in financial markets and economic activities. When housing bubble was bursting, people's expectations changed. Their confidence on the structure products whose underlying collaterals were mortgages from housing bubble collapsed. High-order expectation dried up liquidity of credit market in a second, cutting the source of funding for financial institutions. When one of the companies was first hit by the funding problem, and started to de-leverage positions and liquidate assets, the assets on the balance sheets were priced with a big discount, or haircut so that the value of other companies began to depreciate.
Part of the reason comes from "mark-to-market" accounting, which required financial institutions to write down their assets to current price no matter how low it is. So companies with funding troubles have to sell more assets to get what they used to get. A series of large scale of de-leverage and write-down in hope of obtaining more funding triggered a systemic downward pressure on asset prices and high volatility, making funding even harder. In the absence of functions, liquidity risk, credit risk and counterparty risk were all connected and amplified by a downward spiral, which many economists and regulators have mentioned (Adrian and Shin 2008, Brunnermeier and Pedersen 2008, Dudley 2008, Geithner 2008, etc).
Although the Fed correspondingly launched several other credit facilities early this year in order to inject funding to markets and restore investors' confidence as well, the cost of funding is still high recently as the Libor-OIS spread was at 0.73 bps on July 30, 2008 (see figure 7). The main reason is because of the balance sheet pressure from financial institutions according to William Dudley (Dudley 2008). He said:" Although banks have raised a lot of capital, this capital raising has only caught up with the offsetting mark-to-market losses and the increase in loan loss provisions". In other words, despite the immediate liquidity facilities available, the Fed cannot ease balance sheet constraints materially, and reverse the downward trend consequently. The estimated losses from the crisis can be more than $400 billion according to the paper of Greenlaw, etc (2008). Based on my understanding, the downward spiral in asset prices will not stop until most of companies' over-evaluated assets have been written down and their balance sheets are rebuilt.
Another important determinant is house prices. So far, there is no evidence that housing market will recover soon, the fall of house prices became faster instead. Its annual percentage change in May is -4.8%, down 0.2% from April (see figure 5). Even worse, as home prices continue to decline and credit market is tightening, it is harder for people to refinance loans or sell their homes. The Alternative-A mortgages and prime mortgage, which in common sense have better quality than that of subprime mortgages, deteriorated these days. The delinquency rate of both two rose to 12% and 2.7% respectively in April (Bajaj, 2008). Although they are low relative to subprime mortgages, the speed seems to accelerate. This could be a very bad piece of news since the sum of Alt-A and prime mortgages is more than two-thirds of total mortgage loans, and investors' confidence will be further diminished.
The above is an explicit description of what have been happening since last summer, or the inner loop between housing market and credit market in figure 14. However, it is just a beginning. The aftermath of the subprime crisis is slowly transmitting to the overall economy, or the outer loop in the figure via expectations. As the housing prices have not yet hit the bottom, and the unemployment rate in July climbed to a four-year high announced by Labor Department recently, the economic outlook is very poor. And as negative signals keep sending out from the inner loop, the outlook will continue to be pessimistic for a long period, perhaps the second half of 2009.
In addition, Figure 14 is consistent with the Fed's monetary policy. In the Federal Open Market Committee (FOMC) meeting of August 5, the Fed kept the federal fund rate at 2%, indicating that the Fed slightly shifts its focus from inflation to threatens from badly damaged housing and credit markets, and high unemployment rate. The headline CPI of June touched the highest level since 2000 but the core CPI is stable at 2.4% (see figure 15). As Martin Feidstein points out in August, "it (the Fed) is prepared to gamble that the weakness in US employment and the general decline in economic activity will prevent a wage-price spiral without further increases in the interest rate" (Feidstein 2008b). So inflation is expected to return to comfort zone when or even before the inner loop restores normality.
This is why, globally speaking, U.S. is the only developed country that implements expansionary policies to fight with the recession. By taking advantage of weak dollar, exporters are selling more, helping to sustain overall economic growth. According to Morgan Stanley, net exports accounted for 30 basis points of the 1.8 percent growth in US output over the past year. But a weak dollar may be responsible for high oil prices. Since oil exporters use US dollar to price oil, their purchasing power decline when dollar depreciates. In order to offset the impact, they push the price up by constraining the supply. Figure 16 shows that after a small decline in 2006, the oil price surged to its highest level of $145 per barrel in July, 2008. And during the same time, the US dollar has decreased about 20% to euro. But this explanation is not robust since oil price in other currencies also increased a lot.
Another possible reason could be the low price elasticity of oil according to many analysts, including a report of Interagency Task Force on Commodity Markets (ITF 2008). Low elasticity means that both supply and demand are not very sensitive to prices. When the demand increases, it has to take a quite large price increase to return market to equilibrium. However, the supply and demand maintain a balance these years. The supply and demand of oil in 2006 were 84.28 and 84.8 million barrels per day respectively, and 85.23 and 86.13 million barrels per day in 2007 (IEA 2007). So, it I think will be more convincible if we include the expectation factor. As Martin Feidstein said on Wall Street Journal in July:"... with no change in the current demand for oil, the expectation of a greater future demand and a higher future price caused the current price to rise." (Feidstein 2008a). If oil exporters realize that they can earn more by selling oil in the future, they will store the oil in the ground now, and thus curb the supply.
Therefore, oil exporters' expectations and oil's supply and demand affect each other. This may expain the rise in the prices of commodities like iron ore that do not have active futures markets although such interaction is not transparent. Futures market can reflect such dynamics through trading like hedging and speculation. As we know that futures prices are a consensus of the opinions of market participants, the surge in oil price implies that future oil demand will increase a lot if participants are assumed to trade based on rational expectation.
However, this assumption may not be true. Figure 17 shows that the annual percentage change of oil price jumped from zero to 40% right at the beginning of the subprime crisis last summer. A very likely interpretation to this is speculation. Although many experts such as Interagency Task Force (ITF 2008) argue that the distinction between hedging and speculation in futures markets is not clear, and the supply and demand are the key to oil price, former Fed chairman Alan Greenspan told Financial Times in August that "speculation was importantly responsible for the rapid move up in oil prices in late 2007 and early 2008." He also said that "while the long-term rise in oil prices was wholly a physical phenomenon, financial speculation influenced the timing and profile of price increases." Usually, traders who combine options and futures positions can speculate on price and volatility changes in addition to the direction of expected futures price changes. The recent drop of oil price may be due to speculation on positive expectations like the increase of future supply and a slower depreciation in US dollar.
Also, the change in oil price significantly affects financial stocks, between which there exists a perfect negative correlation these years (see figure 18). And given high inflation, it is a channel linking the inner loop and outer loop in figure 14. To be honest, the origin of high oil price as well as commodity prices is a puzzle to me. What I do know is that although the subprime crisis severely hit the US economy, the high prices spread inflation pressure across world. The outer loop in figure 14 can also apply to other countries. Even worse, under weak US dollar, hot money is forced to flow into rising emerging markets which are already facing high inflation, posing developing countries a more difficult balance between economic growth and high inflation. So Chinese Premier Wen Jiabao urged United States to overcome its credit crisis soon and stabilize its currency when he met with U.S. Secretary of State Condoleezza Rice in this June.
As an economist on the street said, "Time is the only real healer here, giving the markets enough time for price discovery, giving the markets time to identify and quantify all the subprime losses and time for financial institutions to rebuild their capital where necessary." And for me, it will be very interesting and helpful to learn the essence of capitalism from its foundation in Bernanke's bubble laboratory for two years.
Figure 14
Figure 15
Figure 16
Figure 17 Figure 18
Reference
Tobias Adrian and Hyun Shin, January/February 2008, "Liquidity, Monetary Policy, and Financial Cycles", Current Issues in Economics and Finance (14) 1.
Tobias Adrian and Hao Wu, July 2008, "The Term Structure of Inflation Expectations", working paper, New York Fed.
Olivier Armantier, Sandra Krieger, and James McAndrews, July 2008, "The Federal Reserve's Term Auction Facility", Current Issues in Economics and Finance (14) 5.
Adam Ashcraft and Til Schuermann, 2008, "Understanding the Securitization of Subprime Mortgage Credit", working paper, New York Fed.
Arturo Estrella and Gikas A. Hardouvelis, June 1991, "The Term Structure as a Predictor of Real Economic Activity", Journal of Finance.
Arturo Estrella and Frederic S. Mishkin, June 1996, "The Yield Curve as a Predictor of U.S. Recessions", Current Issues in Economics and Finance (2) 7.
Arturo Estrella and Mary R. Trubin, July/August 2006, "The Yield Curve as a Leading Indicator: Some Practical Issues", Current Issues in Economics and Finance (12) 5.
Arturo Estrella and Hao Wu, July 2008, "Term Premiums and the Predictability of Real Economic Activity", working paper, Ner York Fed.
Vikas Bajaj, August 4, 2008, "Housing Lenders Fear Bigger Wave of Loan Defaults", New York Times.
Gurdip Bakshi, Peter Carr, and Liuren Wu, 2008, "Stochastic Risk Premiums, Stochastic Skewness in Currency Options, and Stochastic Discount Factors in International Economies," Journal of Financial Economics, 87(1), 132-156.
Ben Bernanke, March 20, 2006, "Reflections on the Yield Curve and Monetary Policy", speech before the Economic Club of New York.
Ben Bernanke, June 3, 2008, Remarks on the Economic Outlook at the International Monetary Conference, Barcelona, Spain.
Markus K. Brunnermeier, 2008, "Deciphering the 2007-08 Liquidity and Credit Crunch", Journal of Economic Perspectives (forthcoming).
Markus K. Brunnermeier and H L.Pedersen, 2008, "Market Liquidity and Funding Liquidity", Review of Financial Studies (forthcoming).
Bryan Burrough, August 2008, "Bringing Down Bear Stearns", Vanity Fair.
William Dudley, May 15, 2008, "May You Live in Interesting Times: The Sequel", Remarks at the Federal Reserve Bank of Chicago's 44th Annual Conference on Bank Structure and Competition, Chicago.
Martin Feldstein, July 1, 2008a, "We Can Lower Oil Prices Now", A17 Opinion, Wall Street Journal.
Martin Feldstein, August 8 2008b, "The Crisis: A Tale of Two Monetary Policies", P9 Comment, Finanical Times.
Timothy F. Geithner, September 15, 2006, "Hedge Funds and Derivatives and Their Implications for the Financial System", Remarks at the Distinguished Lecture 2006, sponsored by the Hong Kong Monetary Authority and Hong Kong Association of Banks, Hong Kong.
David Greenlaw, Jan Hatzius, Anil Kashyap and Hyun Shin, February 2008, "Leveraged Losses: Lessons from the Mortgage Market Meltdown", Working paper.
Alan Greenspan, July 11, 2005, Response to Jim Saxton's Questions in the Hearing before the Joint Economic Committee on June 9, 2005.
Interagency Task Force on Commodity Markets (ITF), July 2008, Interim Report on Crude Oil.
International Energy Agency, July 2007, "Medium-Term Oil Market Report".
Justin Lahart, May 16, 2008, "Bernake's Bubble Laboratory", Page A1, Wall Street Journal.
Jun Liu, Francis Longstaff and Ravit Mandell, September 2006, "The Market Price of Risk in Interest Rate Swaps: The Roles of Default and Liquidity Risks", The Journal of Business 79, 5, 2337-2360.
Roger Lowenstein, April 27, 2008, "Triple-A Failure", New York Times Magazine.
Roger Lowenstein, 2000, "When Genius Failed: The Rise and Fall of Long-Term Capital Management", Random House Trade Paperbacks, New York.
Frederic S. Mishkin, February 29 2008, "Leveraged Losses: Lessons from the Mortgage Meltown", speech at the U.S. Monetary Ploicy Forum, New York.
Jun Pan and Ken Singleton, September 2006, "Default and Recovery Implicit in the Term Structure of Sovereign CDS Spreads," Working paper.
01.08.2008 pains and gains今天心血来潮，做了一张表，一来揭露一下万恶的，吃人的资本主义，二来预祝2008北京奥运成功。
07.02.2008 Enjoy my life最近中国天气不太正常，上海几十年没下这么大的雪了，古书上说此乃兵戈之象。前天美股暴跌恰逢中国除夕夜，以这种方式迎接新年，是否在预示着些什么。2008年注定是中国之年，一定会有惊喜出现。回顾一下本命年，一切还算顺利。新年伊始，打算开始享受生活。虽然还未到真正享受的年纪，但现在的生活稍纵即逝，在不抓紧就溜走了。尽管当下经济形势一片乌云笼罩，不过找对策，想方法那是头头脑脑们要琢磨的事，I begin to enjoy my life...按照现在流行的说法，我现在坐公共汽车也开始投真币了，存自行车不讨价还价了。我使劲地吃，拉面，龙须面，担担面，方便面，细条的，宽条的。我玩命地吃，一次买十个煎饼果子，让他多放葱花，他就多放葱花，让他少放辣酱，他就少放辣酱，我看谁敢拦着我。
前天和同事们一起去看了New York Giants 的 parade。办公大楼周围平时了无人烟的街道全都被挤得水泄不通。不知从哪里飘来的纸屑满天飞舞。过了许久，百闻不如一见的super bowl trophy 终于在我们的苦苦等待中出现了。随着大巴士在我们面前缓缓驶过，人们不断地欢呼，鼓掌。我想妈祖出巡应该也是这样吧。可以聊表欣慰的是，上次trophy在New York出现是18年前的事了，所以下次再想看到它，必是海枯石烂，沧海变桑田（Giants的fans表打我）。
值此新春佳节，祝各位心想事成，万事如意，赚‘鼠’不完的钞票，享‘鼠’不尽的福。
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
