The EU's bail out

As I mentioned previously, the approach to saving Greece is very similar to what was done to bail out the broken financial system. This weekend a program was announced that lead many people to draw and analogy to TARP: they are spending almost $1 trillion dollars to try to stem the tide of sovereign defaults.  At the same time, the ECB announced that it will engage in credit easing for sovereigns. The difference between credit easing and quantitative easing is that while the European Central Bank might purchase debt instruments that are selling off, they are going to buy other debt instruments to counter any monetary stimulus. This is their attempt to contain volatility. Judging from market movements, it might actually work in the first place.

 

However, this reform does not address the central problem of the Greek and other problematic governments: They are not currently capable of running a budget surplus even if they didn’t have interest payments.  Banks were insolvent but otherwise profitable; these countries look insolvent and unprofitable.  Reducing the volatility will stabilize the markets and help save Europe’s banks that own the volatile debt, but a country going bankrupt is not going to stop going bankrupt merely because volatility is reduced.  Perhaps the sovereign debt should be thought of as more like the US housing market. The US government needed to stabilize the housing market (it is now 80% or 90% of the market according to Shiller) to save the banks, just like the EU needs to stabilize the problem countries.  If the banks holding sovereign debt can earn their way out of the trouble they are in then a few years from now the restructuring of sovereign debt can be more of a political issue than a financial issue.

"Why doesn't some tech guy just arb this?"

A lot of the tech guys I've talked to since the market crash look at the irregularities of a market panic and think that a tech guy looking to make a few bucks can make a program to make a ton of money the next time something like this happens. There are a few problems with that approach.

1. Setting up an running the kind of program that will make money in these situations costs money. Retail investors had really bad liquidity on thursday.  Making money on a rare event that only happens every few years is probably not a wise business decision. 

2. Even if you are good enough to set something up and correctly trade it, you could end up screwed. Nasdaq is cancelling trades on stocks that were 60% away from their "correct" price. So if you saw a stock trading at 40 drop to a few cents, bought the stock for pennies, then sold it for 30 dollars after it started to recover you are in a very difference position than the one you thought you were in originally. Rather than making about 30 dollars and having no position on, you are down 10 dollars for every stock you traded and you still have a short position that you need to close out.

3. Trading extreme drops is just as likely to get you burned as make you money. A stock can drop quite a bit because their drug wasn't approved, or a big lawsuit goes against them.  These types of one off events are probably just as common as the technical crazyness that occured on Thursday.  Maybe the next time markets all drop it is because of a terror attack or a real sovereign default. The programmer who decided to wade a little bit into finance could find themselves badly burned.

The above is also why quant funds specialize in trading generally trade in a style that resembles market making firms rather than a kind of type of special situation trading fund.  High frequency trading is also more likely to generate results that a statistically significant, which quantitatively inclined people generally like.  The quant market maker's job is to get out of the way when stocks start to move in an extreme manner so they aren't the ones left holding the bag.  On a day like Thursday, the problems were probably exacerbated when the quant market makers stood back around the same time people's automatic stops were being hit. Even though there are days like Thursday, the smart technologist interested in trading should probably focus on something a bit more systematic than technical meltdowns.

Panic like it is 2008

The markets took everyone for quite a ride today. The S&P was down 9.3% at 2:44pm before coming back to end the day down only 3.25%. The Vix, a measure of expected 1 month S&P 500 volatility, spiked from 23.8 to over 40, before settling in at 32.8 at the close.  Currency markets also panicked, the EURJPY traded down 4.9% from open to close, at one point trading down almost 8% to levels not seen since 2001. In commodities, crude oil sold off while gold rallied, a sign that there is financial stress that is likely to negatively impact the economy.

So what is causing this craziness?

1. Technical mishaps. There are rumors that a trader with fat finger at Citi sold billions instead of millions of Proctor & Gamble stock. CNN Money's twitter also suggests that something fishy was going on with P&G's stock quotes.  The funny business in P&G wasn't the only technical mishap story. At least one stock traded for only a few cents on the dollar before jumping back up to over $40 dollars. Vanuguard's Small-Cap ETF also took a pounding before closing at relatively normal levels at the end of day. Part of the stock market's extreme mid-day fall could be explained if the stock's sell off caused programmatic trading strategies to hit their stops and automatically get out of positions.

2. Greece imploding: The Greece protests are making market participants wary of sovereign risk in the rest of Europe. The striking Greek citizens appear somewhat crazy - Greece's primary deficit is over 8%, so even if they wanted to stick it to the financial markets and default on their debt the government would have to implement austerity plans. With this is mind, their protests are a mass protest of reality. Even worse, their striking makes any hope of a Greece recovery even slimmer than it already is. The country can't earn money to pay off the debt when their country isn't working and the tax collectors refuse to collect taxes. The best case for Greece is that the protests are some type of cathartic reaction, and the citizens will get back to work after expressing their anger. Today's market movement suggests that the market didn't buy this story. Furthermore, the money for the Greece bailout has not been 100% secured yet, so rumors about Germany not approving the money may have also scared the markets.

3. Contagion:  The Greece bailout protests also give investors a view into the type of pressure other European countries are facing if they try to implement austerity measures. The other countries will have to implement the plan with a situation far worse than Greece's. For one thing, Merkel's party is probably going to be punished over the Greece bailout in Sunday's Nother Rhine-Westphalia election, reducing the probability of other countries getting a similar bailout.  Furthermore, the current administration in Greece is still viewed more favorably than the opposition party, which lied about the deficit and the level of debt. Greece is also implementing the austerity measures in return for assistance, while other countries such as Spain (Which is too big to be bailed out as it is) will have to implement them merely to avoid disaster and without the cover of a hated opposition party. The worry about Spain's ability to repay their debt has been rising, the spread between German and Spanish bonds started the month at 101.3 basis points and has widened to 162.8 basis points.  Worries are so intense that Portugal's 5 year CDS spread is now trading around 453, where Greece's CDS was in mid April.

Without the Greek problems and the contagion worries, the technical issues might not have been that big an issue. However, in the context of these structural problems, the technical glitches caused quite an interesting day in the markets.

City size and Voting Patterns

Some people think that large cities automatically imply liberal cities, while smaller cities are more conservative. This type of analysis ignores the conservatism of Oklahoma City and the liberalism of Berkeley, California (or Gary, Indiana). Using data from the 2004 election and population size data, both from the Bay Area Center for Voting Research report on “The Most Conservative and Liberal Cities in the United States” we can see how much city size moves a city towards being liberal or conservative.

The data is restricted to cities over 100k population, so this might not catch the “small town” effect. Some comments on the drivers of this relatively weak, yet significant, relationship.

  1. Demographics matter more. African American and to a lesser extent Hispanic populations (excluding Cubans) will vote liberal regardless of whether or not they are in a smaller city.
  2. Cities with older populations will be more likely to be conservative.
  3. Large cities tend to attract young single people, who are more likely to be liberal while the married couples in the suburbs lean more conservative.

The difference between rural and urban areas is larger than the above chart suggests. According to CNN Exit polls, Obama won big cities 63% to 35%, while he lost rural areas 45% to 53%. These numbers are unadjusted for the racial composition of these areas. Still, looking at these factors is more enlightening than the typical “red state/blue state/swing state” approach.

What Caused the Slacking?

Tyler Cowen and Arnold Kling both point a paper by Philip Babcock and Mindy Marks on the declining number of hours students spend studying per week.

After accounting for framing, we observe statistically significant declines in study time of about 8 hours per week between 1961 and 1981, about 2 hours per week between 1988 and 2004, and about 10 hours per week between 1961 and 2003.

The out of class weekly study time for full time college students decreased from 24 hours in 1961 to 14 hours in 2003. The surprising thing about this study is that the bulk of the decline occurred between 1961 and 1981 instead of between 1988 and 2004 during the time when personal computers became really useful for the majority students.

The paper is pretty thorough and finds the decreased study time occurring across different majors and at all prestige levels. However, two possible theories for the change in study habits are incorrectly dismissed: increased female student population and more students working part time to pay for college. These potential causes are dismissed because female students study more than male students, and the hours spent studying by full time students also decreased.  The key aspect here is that both of these phenomena give students the ability to study less and still maintain their relative ranking within the college.  The top quartile student only has to study a little more than his classmates. If the median students now have better things to do than study (dating, a necessary part time job) then the top quartile students can still easily retain their class ranking despite a reduced study time. 

Another explanation is just that the culture of the baby-boomers was not that interested in studying, and the trend was self reinforcing for the relative ranking reason mentioned above.

On a tangential note, if students are graduating while working 10 hours less a week then catching up to the knowledge frontier probably isn't a good explanation of why younger scientists are less productive.

Saving Greece vs. The Finance Bailout

A lot of people seem to think that the way to save Greece is to treat them the same way financial companies were treated after the fall of Lehman Brothers, by giving them temporary freedom from the market. With financials, the freedom from the market was given in numerous ways.

  1. The discount rate was lowered to give banks access to funds at lower interest rates
  2. The Federal Reserve provided short term liquidity to banks by increasing the term of overnight loans to 90 days, creating the Term Auction Facility to allowed banks to borrow Treasuries against less liquid paper.
  3. Market participants were not allowed to be short financial companies for a period of time.
  4. TARP gave money to banks at rates better than they could get on the open market.
  5. Changes in accounting rules allowed banks to avoid reporting their losses.
  6. Banks could exit their AIG trades at favorable marks, giving them free money.

Since this seemed to work for the financial crisis, many of the same ideas are still around for dealing with Greece’s problem.  For one thing, there are people calling for the heads of speculators betting against Greek bonds by buying credit default swap protection on Greek bonds that they do not own.  Now, it looks like an aid package is going to be passed after the IMF and Greece work out a three year program. The main idea is to prevent Greece from having to go to the markets.  They are trying to get the package resolved before some Greek bonds mature on May 19th.  Once the package is resolved, the hope is that Greece won’t have to go back to the market for three years; the whisper number is that 120 to 150 billion Euros will be lent to Greece. The idea is that Greece will fix itself without having to deal with excess pressure from the market that raises their debt interest payments and drives them into a deeper hole.

Using the financial companies as a template for fixing Greece misses a fundamental difference: Greece is producing a large deficit while finance companies were profitable despite their insolvency. Protecting banks from the market worked because the banks could earn their way out of the hole that they were in. In some respects they were more profitable than normal because two of their competitors, Bear Stearns and Lehman Brothers, were basically taken out of the market so the illiquidity of the market provided huge rents to established market makers.  Greece is in the opposite situation. It is losing money and its GDP will take a large hit after austerity measures are implemented. Without a crisis to spur them on, it is questionable whether or not the unions will stop striking. When some of these unions include tax collectors, the Greek fiscal situation is only going to deteriorate further once the bailout puts the immediate crisis on hold.

Greece is still going to be in trouble, but maybe in the three years without a bailout the European banks holding Greece’s debt can earn enough money to prevent the inevitable default from driving them into bankruptcy.  More worrisome, if Greece’s situation doesn’t improve this could be bad for Portugal, Ireland and Spain. Their cost of funding is rather correlated to Greece’s CDS spread, especially for Portugal. If Greece’s CDS spreads get worth then their funding costs continue to go up and there is a possibility of a negative reflexivity loop where their cost of funding goes up so their fundamentals go down which causes their funding costs to go up and so on.

More on Dividends and Capital Gains Taxes

Earlier I suggested that high dividend stocks might outperform low dividend stocks in the wake of capital gains tax increases. This analysis didn’t properly account for changes to the dividend tax rate. According to the Wall Street Journal, the Senate Budget Committee passed a fiscal 2011 budget resolution that includes plans to raise the top tax rate on dividends to 39.6% from 15%.  Adding on the 3.8% surcharge on investment income from the health care bill raises the top rate to 43.4%. This change would completely counter any relative benefits high dividend stocks get from a higher capital gains tax.

Looking at the American Human Development Index

Bryan Caplan points his readers towards the intra-US Human Development Index. This Human Development Index is a measure that gives equal weight to life expectancy, educational attainment/school enrollment and median earnings.  Bryan Caplan’s critique of the very similar international version of the HDI is that it gives far too much weight to the educational measure and far too little weight to measures of GDP, which can only contribute a fixed amount to the scale.  With this fixed scale, rich countries are unable to improve their HDI scores via getting richer, even though this would obviously improve the lives of the population.  There are also good arguments against the absolute level of school enrollment being correlated to quality of life past a certain point. When looking at the international HDI Bryan Caplan concludes 

Scandinavia comes out on top according to the HDI because the HDI is basically a measure of how Scandinavian your country is.

However, the breakdown of the USA HDI is rather interesting. For one thing, we see that Asians consistently outperform Whites, who outperform African Americans and Latinos. Both of these groups are generally ahead of Native Americans. The report gives a breakdown of how well each ethnicity is doing in each state.

For Asians: The highest index score is for Asians in New Hampshire, the lowest Asian score is for those in Louisiana.

Potential Reasons: New Hampshire is relatively wealthy, and yet is less likely to be a first destination for immigrants to the extent of other states such as California and New York. The high score of Asians here might be a measure of second generation Asian success when the averages aren’t brought down by new immigrants. The low score in Louisiana is probably because it is relatively poor and potentially more likely to be racist.

For Latinos: They score highest in New Jersey and the lowest in Alabama.

Potential Reasons: New Jersey has a relatively high amount of non-Mexican Hispanics as well as a large Hispanic population relative to the population, which may bring up the average score of Latinos in this state. In Alabama, already a relatively poor state from a per capita perspective, there is a particularly low ratio of Hispanics in the population and many of them are both young and Mexican, lowering their average score.

For African Americans: The highest score is in Maryland and the lowest score is in Louisiana.  The African Americans with a job in Washington DC might be likely to be living in Maryland, which would raise the pay and educational attainment of the average population.

Native Americans: They score highest in California and lowest South Dakota.

Potential Reasons: South Dakota is not relatively wealthy or much of a tourist destination to begin with and the city of Deadwood legalized gambling, which competes with Indian casinos.

For Whites: The highest White score is for those in Washington D.C. while the score is lowest in West Virginia.

Potential Reasons: Washington D.C. has a large amount of government jobs which both pay more than private sector jobs and employ a large amount of people with humanities PhD’s, which are not highly valued outside of academia or governmental type agencies. This raises their median earnings score and their educational attainment score. Education is correlated to longevity, so this score is higher too, though educational attainment is the main outlier. The relatively large amount of colleges in D.C. should also help the scores. West Virginia has historically been the home to poor whites

More on Greece and the Markets

Greece’s troubles are getting worse, as S&P downgraded their rating to junk. Given that I’ve previously looked at the impact of Greece’s troubles on markets, I thought I would update some of the charts to see if the relationships still hold.

 

At the epicenter of the crisis, the Greek stock exchange has been massively hit by Greek troubles. The National Bank of Greece, now the second largest component of the index (Having fallen behind the Coca Cola Hellenic Bottling Company), has fallen almost 45% year to date.  There are many reasons for investors in Greek banks to be worried. First, these banks tend to hold a lot of Greek debt, which has been rapidly falling in value. Greek bonds with a duration of 10 years have seen their face value fall over 20% in the past two months alone.  Second, the tail risk of Greece leaving the Euro is becoming less tail and more real. If Greece leaves the Euro, there will almost certainly be bank runs on Greek banks as depositors try to prevent their accounts from being devalued.  Finally, even if crisis is averted it will be due to the imposition of extreme austerity measures by the Greek government. Cuts in spending and tax hikes will depress the economy and hurt the earnings of all Greek companies, including financial institutions.

Greek CDS & Athens Stock Exchange

 

The relationship between the S&P 500 vs. Eurostoxx in dollars seems to be holding up quite well. The euro has also been taking quite a hit.

Greek CDS & SPX/Eurostoxx (USD) & USD/EUR

 

When looking into which sectors should be most effected, banks seem like the obvious choice.  European banks should be hit harder than US banks for a few reasons. [Edit: The most obvious reason European banks hold more Greek debt than US banks is that banks can repo the bonds of European Sovereigns with the ECB.]  Due to home bias, European banks have larger investments in Greek bonds than US Banks. Home bias doesn’t quite explain their larger holdings; they also have a search for yield mentality. This search for yield is part of why European banks were buying subprime bonds from Goldman Sachs (Although IKB was also betting on the housing market going up in 2007), and this is why they currently hold a lot of Greek debt. According to BIS estimates provided by the Economist, the non-Greek euro area Banks own approximately 58% of the total outstanding Greek debt held by foreigners while US banks only own 5.3%. Since these banks are still recovering from the last financial crisis, the impact of a Greek credit event could be devastating.

S&P 500/Eurostoxx (USD), US Banks/Euro Banks (USD) and USD/EUR