From bad to worse?

The austerity measures in Europe are starting to bite but not in the right way. As governments impose painful belt tightening measures, recent statistical figures  provided by Eurosat, are pointing to a further bloating (degradation) of debt to GDP for the majority of Euro zone members. That certainly does not bode well, considering that the objective of austerity in the first place was to reduce the debt, not increase it! It could be the result of growing deficits and very weak or negative growth. It could also be that there is a sort of lag effect at play and that, over the longer term, there will be a positive impact on the debt. If this the case, it necessitates time to work, or patience, a commodity that respective governments are rapidly running low on. Country after country, the authorities are switching to crisis mode as the public become more vocal regarding their displeasure on the way things are being handled. Union is giving way to nationalism and the harmony of the past is disappearing.
Europe is in trouble, that we know, but maybe more worrying are the signs that things may be getting worse.

Changes in govt debt to GDP ratio, Q311 - Q411 (Eurostat)

From "dark pools" to "murky waters"...

Are "dark pools" a by-product of the singularity effect on markets?

As computing technology advances and as communication gets closer to being instantaneous, the whole landscape of investing is being impacted. This was first observed in a significant way with the so called "flash crash" of 2010, a glitch in computerized trading algorithms that, for a fraction of a second, sent the Dow tumbling down 9% before recovering.

The exponential growth in computing power and its increasing ubiquitousness in the financial markets are posing a threat to the more traditional approach to investment management. Rapid-fire traders, assisted by computing power running complex algorithms are able to second guess and place orders before your average investor can, thereby reaping the benefits of "first come, first serve". This is proving to be problematic for a whole range of investors that are ending up buying high and selling low.

Dark pools basically provide a way for investors to trade stocks away from public stock exchanges and, more importantly, away from the prying eyes of opportunist rapid-fire traders. Blackrock, for example, is in the process of setting up an "independent" fixed income trading platform for its clients. But even dark pools, it seems, are not immune from the clutches of the singularity effect, and, unfortunately, not always in a benign way.

According to a recent SEC finding, a firm called Pipeline Trading Systems LLC, that was running a so called "dark pool" for its clients, was actually using another firm it had created as the counter-party to its client trades. Even more revealing was the fact that the firm used sophisticated algorithms in its trading platform to outsmart the clients, thereby reaping nice profits from the trades. In other words, far from being shielded from the damages of high frequency traders, the "dark pool" the clients were trading in was nothing more than a mirage. This is unfortunate but could also suggest that possibly the only effective way to counter this trend would be to employ more computing power. With all the emotional bias shortcomings to investing that have become even more evident with the financial crisis, it might be time to rethink the whole approach to investing.

Is Spain the next Greece?

These are dire times for the Eurozone because, despite the positive effects of the LTRO, there doesn't seem to be any follow-up going on. The same medicine is being applied, i.e. forced austerity with no relief in sight. Even more worrying is the signs that the Greek triggered contagion seems to be spreading to the larger "peripheral" economy of Spain and this is where the real troubles begin. The country's fiscal balance and public debt were amongst the healthier ones of the zone when Greece's troubles first surfaced in 2010. Spain entered the crisis as a result of the real estate bubble that was being fuelled with cheap lending from core members. Unemployment is already rampant, at levels comparable to the "great depression" era of the 1930's and the austerity measures are only contributing to deepening the slump.
The graph above clearly shows that the LTRO effect is running out, accelerating the growth of the debt burden through higher borrowing costs. Unlike Greece, Spain's economy is significantly larger making it significantly more difficult to bail it out. If the Eurozone cannot afford having Greece fail for reasons of contagion, imagine the risk that Spain represents!

A novel approach to portfolio construction...

The traditional approach to building an investment portfolio with the aim of achieving a target wealth level for a set date in the future (typically retirement date) is relatively straight forward and involves discounting the future value of this portfolio to the present using an estimated discount rate which comprises a somewhat rough estimate of the expected annual growth rate of the portfolio. The major flaw with this approach is that the rate of return of the portfolio is very difficult to estimate, especially in the current volatile environment where visibility is relatively poor.

A far more intuitive and robust proposal would involve splitting an investment portfolio into two segments. One segment would be designated as the exposure that aims for capital preservation. This exposure contains investments that have very low risk and very low probability of experiencing drawdowns. The size of this segment should reflect the absolute strict minimum amount of wealth that the investor will need once he or she reaches retirement. The other segment comprises of the significantly riskier portion of the portfolio that the investors could afford to lose entirely but that should provide steady growth of the capital over the longer term.

You may wonder what the difference between this approach and a classic asset allocation portfolio comprised of a diversification between bonds and equities may be. The key difference is the the so called low risk allocation is there purely to preserve capital, it won't grow on its own but will increase in time through the reallocation of capital from the riskier segment, assuming that this segment does actually grow. The classical approach runs the risk of experiencing a sharp increase in correlations similar to 2008 when practically all asset classes were losing value simultaneously. Drawdowns in this novel approach should, instead, be limited to the "risk" portion of the portfolio.

The challenge in implementing this is twofold:

1. How do you ensure that the "capital preservation" allocation does its job in a world where capital losses may be incurred even with money market deposits, through the uncertainty regarding counterparty risk (Lehman's being the recent example)
2. The capital transfer from high risk to the low risk segments assumes that the high risk segment will grow through time. Nothing is less certain if we look at stock market performances over the more recent past!

Still, despite these challenges, the novel approach does provide a solution that is far more adapted to the current market environment and thus has a better chance of succeeding over the longer term.

Ripple effects...

A deal was struck (if we can call it a deal) last week between Greece and its private bondholders where up to 80% of them "agreed" to exchange their toxic debt against new ones that are thought to be less than half the face value. I guess the rational behind the acceptance was that it was a choice between getting something or nothing.

The more interesting question to ask is if this deal will trigger payments from credit default swap insurance on Greek bonds. Normally it should because the agreement is clearly akin to a technical default. If the body that decides on such matters argues that it is not a default, they run the risk of significantly damaging the credibility of the CDS markets, causing even greater burden on the borrowing of the other Euro-zone economies that are struggling with their debt.

If, on the other hand, they consider this action as a default, there is still great uncertainty as to how it will unfold. Just as in the Lehman and Dexia debacles in 2008, nobody really knows what the counterparty risk really is. We may have an idea of the amount of exposure and the main counterparties but we don't, for example, really know the counterparty to those counterparties.

The point is that in either of the two scenarios, there is bound to be ripple effects and it is difficult to estimate beforehand what the outcome of those ripple effects are likely to be.

If reason prevails (not always a given in financial markets), the governing body will rightly recognize the deal as the equivalent to a default. Greece has evidently defaulted on its debt, the question is, where will it spread next?

What is an equity index really measuring?

Equity indices are designed to measure the performance of a specific sector, market, region or even the entire world. They are meant to provide an indication of the type of returns you can "expect" from equities over the longer term but they also tend to suffer from some serious shortcomings that can have serious consequences.

Most popular indices for example only capture the price performance of its constituents typically weighted by capitalization (S&P 500) or just price (Dow Jones Industrials). What they don't capture, however, is the impact of dividends, especially when they are reinvested. Total return indices do exist, but they are unfortunately much less common and therefore rarely used and so investors are much less aware of them.

Even more rare are indices that are adjusted for inflation. That type of information is just not published but the impact can be very significant.

To show just how significant the effect of dividends and inflation can be on the return of an index, researchers in the U.S. did the calculations on the Dow Jones Industrials index and here is what they discovered:

If dividends were included since the inception of the DJI in 1896, it would have a value of more than 1,300,000 which is more than 100 times its recent high of 13,000!

If it was adjusted for inflation over the same period, it would shrink to a value today of around 500 which his about 96% below its actual current value.

Finally, if we were to combine the impact of reinvested dividends and inflation, the figure would be around 47,000 or almost 4 times its current value! In other words, the impact of reinvested dividends seems to be substantially greater than the impact of inflation (assuming inflation is correctly measured of course!)

All this to say that it is important to be aware of what exactly is being measured (or not) when looking at an index. Just as in the differences between price weighted and capitalization weighted indices, the effect of including dividends and/or inflation can lead to substantial differences in the results.

Are ETFs boosting correlations?

I came across a recent study that points out to a sharp rise in correlations between stocks right about the time when the ETF market has experienced exponential growth. Is it possible that the popularity of ETFs has led to a jump in equity correlations? It does seem plausible if we consider the way ETFs trade.

When you want to purchase an ETF, you usually go to the secondary market where buyers and sellers congregate. When the purchase amount is significant, however, it needs to be created through the primary market and the way this is done is through one of the designated brokers that deal directly with the sponsor in the primary market. The broker uses the investment cash to purchase a basket of stocks that reflect all the constituents and their respective weights of the underlying index. This is then exchanged against an equivalent number of units of the ETF. In the event of a redemption, the exact opposite happens, whereby the broker exchanges units of the ETF for an equivalent basket of stocks representative of the index. Now imagine a bear market scenario where a whole lot of redemption activity is going on. In the case of a broad market index ETF, the broker will have to liquidate a large number of stocks right about the same time. In other words, all the constituents of the index will be experiencing selling pressure at around the same time which means an increase in the correlation between them.

This estimated general increase in correlations has led to a situation where you need to hold a larger number of stocks to achieve an optimal level of diversification. It also further emphasizes the importance of ensuring a broad level of diversification between different asset classes.

Just about right...

The Fed has kept monetary policy aggressively loose and intends to do at least until sometime next year. Of course, this will depend on how the economy unfolds. Loose monetary policy has been the necessary medicine to keep the economy afloat, ever since the housing market bubble popped in 2007 but it has largely been ineffective in part because the Fed found itself in a "liquidity trap". The transmission system between Fed policy and bank lending is broken but there are signs that it is healing.

Monetary policy, in "normal" times can be a potent tool, which is why it needs to be wielded carefully if price stability is to be maintained (one of the two mandates of the Federal Reserve). One tool that can help determine whether the target rate is set "correctly" is what is known as the "Taylor Rule" named after the U.S. economist John Taylor. In its basic form, the Taylor Rule is a linear equation that determines what the Fed target rate should be. To do this, it takes the current rate of inflation and compares it to the "optimal" rate of inflation. It does the same with growth by comparing current output or GDP to "potential" GDP. Basically, the required target rate changes if either of these two factors are currently above or below their optimal or potential levels.

The graph above compares the actual Fed Funds Rate (white line) to that of the Taylor Rule Estimate (blue line). It is interesting to note that overall, the actual target rate and its estimate seem to be relatively well correlated and that currently the Taylor rule estimate seems to agree almost fully with the target rate. In other words, the estimate seems to fully support Fed policy.