High Frequency Trading

I recently was asked about HFT and decide to include my comments here.

High-frequency trading (HFT) accounts for about 70% of all U.S. equity trading volume. I disagree with some that this has turned our stock markets “…into an enormous betting system…” though. All HFT requires algorithmic trades (AT) but all algorithmic trades are not high-frequency. And neither algorithmic trades nor high-frequency I believe pose a significant problem.

I do believe in the Efficient Market Hypothesis (all information is available and therefore there truly is a “random walk” for stock prices) -- long-term. Ah, but what about the distortions we have seen in short-term trading? Yes, that does, indeed, exist so I will highlight that short-intervals of time (milliseconds, seconds, minutes, maybe hours) do not necessarily follow a random walk and that is a result of AT and HFT.

However, HFT provides liquidity to the markets (provides a buyer or seller, as needed) and reduces overall trading costs by narrowing the spread between the bid and ask prices. The price all investors pay is in some additional volatility (short-term, usually measured in minutes, hours) yet not a concern for long-term investors. Offsetting that volatility are somewhat discounted trading costs.

I remind myself that what happens over an hour, a day, a week - even years, for me and my clients - should not affect an overall allocation strategy. It is personal. What is happening in one’s life that would possibly change that strategy rather than the economy or the style of electronic trades.

I read an interesting comment that HFT (that, is the speed and the frequency) are not the concern, but only when manipulative strategies are used with it. Most HFT is a result of large volume but very small trades (except when errors occur - like May 6, 2010 - or someone is truly trying to manipulate the markets, hopefully they get caught). AT and HFT provide a market for every buy and sell transaction - very quickly - and actually adds to price continuity. I really believe that is true. 

When I process a stock transaction it is filled immediately. Whether bought by the broker for their inventory or by the HFT firm that makes a fraction of a cent filling my order, they are then turning around and reversing the transaction with someone else.

"...HFT firms make their money in the buy and sell part and lose money when they hold a position purchased (or sold) for more than 5 seconds...." (SSRN article) Their desire is to have a net zero position by the end of the day. You may have noticed that most trades occur at the market opening and then at the close. 

If you would like to read or research more about this I recommend: The Social Science Research Network (SSRN) at:   www.ssrn.com (you can register and have access to great research papers at no cost for just using the website articles).

There is much research on this topic, diverse opinions and a continuing desire by academics to fully understand the ramifications of this on the markets.

Expected Return is What Counts

When prices go down buy and hold. Why? Because of future returns...

Here are two concepts about "expected return" quoted from economists Fama and French at their website: www.dfaus.com :

"...in the short run the number of shares outstanding is fixed. As a result, changes in risk cannot affect the aggregate portfolio of all investors; you cannot reduce your equity position unless someone else is willing to increase his. Changes in risk can, however, affect price. When risk goes up I expect prices to fall and expected returns to rise..."

AND

"...the onset of high volatility should be associated with price declines that increase expected returns going forward (to compensate investors for the higher volatility), and the onset of a low volatility period should be associated with price increases that lower expected returns going forward. As a result, if you bounce in and out of the market in response to variation in volatility, you are likely to be in when expected returns are low and out when expected returns are high..."

10-year Record

According to Morningstar's Ibbotson records (and stated in the March 2011 Financial-Planning.com magazine article written by Donald Jay Korn), "...the best 10-year record for large-cap stocks...occurred in the post-World War II boom of 1948-1957..."

This intrigues me since the annualized return of that 10-year period exceeded 20% at a time when in 1944 the government debt as a percentage of GDP was a whopping 73% (see my blog entries on the Economy-General). This number gradually continued a decline to only 18% several years ago though it has been on the rise to around 25% today.

10-year Periods of Negative Returns

Since 1871 there have been 14 ten-year periods where stocks experienced negative returns.

We just lived through one such decade. That is 10% of the time.

According to analysis by Jeremy Siegel, the real stock market returns in the decade following every one of these 14 periods averaged in excess of 10%. Past performance is not an indicator of future returns, however, history does have lessons to teach us if we are willing to listen.

Simple International Investment

Keep in mind that 40-50% of the revenues from S&P 500 companies (those multi-national companies whose names we all know that represent 80% of this index) come from their sales to overseas countries.

Wow!

Questions Posed to Kiplinger's Magazine Part I

In the March 2010 issue of Kiplinger's magazine, readers posed questions and the responses were overall pretty reasoned but here is one that I had trouble with:

"...Stocks still terrify me. How can I get back in the market with less risk?..."

Kiplinger's response was first to invest through a merger fund and second to consider a balanced fund (that owns stocks and bonds) and finally, regardless of fund chosen, to buy in gradually.

Now for my "real" answer:

Get over it. Stocks are risky. Short-term, that is. They go up and down.

Long-term, however, stock returns are not your worry. Inflation and taxes are what should be terrifying you. Don't invest in the market with a short-term outlook.

You can not lessen risk (the original question) by buying a merger fund (one of the riskier styles) nor by buying a balanced fund (if bond interest rates go up, then that portion of the fund will go down).

And, do not dollar-cost average into stocks if you have the funds already. That is useful for payroll deductions when you do not have the money readily available but makes no sense if you happen to have a stash. Diversify immediately. That may require some lump-sum purchases into stocks if none exist in your investment portfolio.

Understanding market volatility certainly would help. Time horizon is so very important as well as what your income needs are from your investments.

Downturns and Recoveries

Much is written about steep drops in the stock market and the eventual recovery. In basic terms, it is fear and greed that drive the market and as Stephen Savage wrote in Financial-Planning.com magazine (September 2008, over a year ago), "...the sense of fear that magnifies perceptions of risk...causes stock prices to discount more than fundamentals would dictate..."

In October of 2007, the stock market (as valued by the DOW at over 14,100) was not over-valued and at dizzying heights as it was in March of 2000 (when the DOW was at over 11,000). The DOW value is relative to earnings and growth projections.

What matters is patience and riding out severe discounting of the market as well as understanding of the severe markups in the market for future growth. In March 2000, the P/E (price-earning ratio) was well over 30 and the recovery (months to recover from the trough)took 51 months. In 1980-2, the P/E ratio was around 8 (do realize that this is fictitiously low also because inflation was so high and distorts all the numbers) but the recovery took 2 months.

In October 2007, the P/E was hovering above its historical average of 15-16 but it was not in the stratosphere of the year 2000. The 1920's also was a crazy time of exuberance before the Great Depression and has more similarities to the 1990's than to the 2002-2007 period before our current Great Recession.

The months it takes to reach the bottom (the trough) also portends, historically, how long it takes to recover. It took 30 months to go from March 2000 highs to the low on 10/9/02 and 51 months to recover. From October 2007 until March 2009 was about 17 months (if, indeed, we do not see that low again - as there are no guarantees that we are through with this). That is a pretty quick decline and one reason why everyone felt it more. The 2000-2002 decline was 46% and the 2007-2009 decline was 52% but it felt worse.

Anyway, since 1950, there have been 13 significant drops like this. Recoveries have averaged from 2 months to 51 months. Only time will tell.

A Little DOW JONES History

The largest one-day percentage gain in the DOW index, 15.34%, happened on March 15, 1933, in the depths of the 1930s bear market.

The largest one-day percentage drop occurred on "Black Monday", October 19, 1987, when the average fell 22.61%.


This may be a good time to reflect on excerpts from an article from last year by John Reese (3/5/08) at MoneyCentral.com:


"...adding Bank of America and Chevron and dropping Honeywell and Altria...recently..." (early 2008)

Per the Dow Jones Release: "...B of A was added because financials were underrepresented in the index, and Chevron because the oil and gas industry's importance to the world's economy is growing..."

What timing.

In summary, author of the March 2008 MSN article, John Reese, states:

"...If the 30 stocks that make up the Dow don't look appealing at a given time, that doesn't mean you should avoid the market altogether. With thousands of stocks out there to choose from, you can always find good values -- whether the Dow looks good or not..."

So remember:

First, Bank of America was not even in the DOW JONES INDUSTRIAL AVERAGE in 2007.

Second, The DOW is price-weighted, meaning each company makes up a fraction of the index that is proportional to its price not market capitalization (price times number of shares).

Third, since the DOW is a price-weighted index and financials and GM have nose-dived, their large effect on the DOW's down-side may be less in the DOW's eventual recovery which may come from the non-financial companies. No guarantees just an interesting thought. GM is the only auto manufacturer in the DOW - not Ford or Chrysler or any others.

Fourth, Dow's stocks have an average market capitalization of $147 billion, versus $28 billion for the S&P 500. They are "big" companies but it is price (with adjusting factors) not market-weight that makes the difference. (statistics change every day but this is an idea of size)

Fifth, (per Wikipedia history) "...the "industrial" portion of the name is largely historical—many of the 30 modern components have little to do with traditional heavy industry. At one time it had 9 railroads, 2 Industrial stocks and GE (the only company that has survived to today)..."

The point: the composition of the DOW changes from time to time. It may be in for another change after this avalanche.

Sixth, Microsoft, Intel, Home Depot and AT&T were added in 1999 replacing others and Pfizer and Verizon were added in 2004 replacing other companies. This has an effect on the DOW index and should be watched.

Stock Market UP 20% !!!!!

Yes, you will not see this in the media anywhere but the DOW Jones Industrial Average has increased 20% in 3 weeks. (The media reports the month of October as the worst since 1987 to further discourage you).

From an intra-day low (never reported this low by the end of the day but during the day) of nearly 7,700 (near or around October 9th) to Friday's close on October 31st of 9,300+.

I have no clue whether we are headed lower than 7,700 some day in the future but as of now, those who did not sell low have recovered 20% of their losses in 3 weeks. The return on safe money markets and CD's during the same period is less than a 1/2 of one percent and probably less than that. Of-course, those "safe" money returns are negative when inflation is considered.

Let's Look Again at History

A recent article (September 2008) in a respected financial planning journal (http://www.financial-planning.com/) by Stephen Savage entitled "Anatomy of a Recovery" highlighted the recovery times if, mind you, you do not give up and move assets from stocks at these low values to other fixed income areas. Recovery times are much longer if you "bail".

He states, and I agree, "...typical of all downturns: the sense of fear...magnifies perceptions of risk and causes stock prices to discount more than fundamentals would dictate..."

Here is some perspective:

In 1968-1970 the stock market took 18 months to lose 36%. It recovered that in less than one year.

In 1973-1974 the stock market took 21 months to lose 48%. It recovered in just over 4 years - quite a bit longer but the recovery did come.

In 1987 the stock market took 4 months to lose almost 34% (where we are as of October 9th as I write this and that took 12 months to lose this amount). In 1987, that 34% loss was recovered in 1 and 1/2 years.

In 2000-2002 the stock market lost 49% over a very long 30 months (almost 3 years). This recovery also took the longest: 51 months to recover by December 2006.

HISTORY LESSON: recovery does occur (that is, historically, of-course). Smile.

DOW Down 350 points in 24 hours

The DOW is starting with 1o thousand again and is at 10,850 at 10:00 AM on Tuesday July 15th when it was at 11,200 at 10:00 AM on Monday July 14th - a 350 point decline.

I am sure there are some who are wondering when this despair will end. When the bottom is finally reached. When recovering begins. Me too? I have no answers. No one does. Sorry.

I have no answers except "Ice Cream". Not: "I Scream". You see, when ice cream goes on sale for half-price or even just a 23% discount (like the DOW is right now) the freezer usually gets stocked with more (unless you really are being careful with the weight issues).

If you buy more when stocks are on sale does that mean that they won't continue to go down? No. However, from 1965 through 1984 the DOW seemed stuck. You know what number it could not seem to break-through? 1,000. Yes, there is no zero missing. One Thousand.

So, personally, with a long time horizon, I am not concerned about this recent market turmoil. It will pass and we will recover from this decline.

I Am Scared Too !!

I am scared. Not of the stock market decline but of the doom and gloom so often reported, as you may be, that my emotions do signal me to do something. Something. How can any human being not be so affected? It is natural.

USA TODAY's headline for Tuesday July 1, 2008 reads: "Stocks lose $2.1 TRILLION DOLLARS - WORST SINCE GREAT DEPRESSION"

I read that and thought that this cannot be correct based on my information and studies of historical returns but my emotions got the better of me. After reading the article I found that, in reality, $1.4 trillion of the $2.1 trillion was lost in the month of June alone. Yes, a $1.3 trillion loss for the single month of June may be the worst since the Great Depression but that is a very different message.

The headline was very misleading but it did get me to do what headlines do, namely, to read the paper. And, there was no mention of what happened after the month of June. The Great Depression saw the stock market lose 80% of its value over an almost 3 year period (so we are not even close to that mark as the headline mislead me). Yet, in 1932 the stock market had its best gain ever of 42% - rarely mentioned.

Here are the facts:

We are down 10% for the month of June and nearly 20% since the high in October of 2007 - only 262 days ago. A 20% decline or more occurs, on average, about every 5 years (not every 70 years) and the average decline is about 26% and lasts for 11 months (363 days). The last time was 2000-2002 and it was longer than the average. We are just about right on time for this decline.

In the Washington Post on Sunday June 29, 2008, a well-known investment strategist for a large mutual fund company is quoted as saying that we may be in for a long and deep recession the likes of which we have not seen since WW II. Maybe? Maybe not?

With all of this pessimism, though, I am glad that I have, and have prepared for my clients, an Investment Policy Statement that reminds us why we might change the original investment allocation strategy. Entering a bear market is not a reason to change my strategy and selling stocks is assuredly not the action to be taken during these emotional times. What is? Possibly rebalancing by buying more stocks but that depends on each individual situation.

Ned Davis Research shows that there have been 33 bear markets since 1900 or about 30% of the time which means that stocks are going up twice as often as they are going down. During a snow storm it may seem like it will never stop but it does. This decline, too, will pass. Don't let your emotions control your investment actions.

Stock Market for June 26, 2008

As I write this at 1:00 PM on Thursday June 26th, the Dow Jones Average has gone down about 235 points to 11,576. Where it goes from here I will not speculate but I will make some comments:

At 11,576 this is a 2,600 point decline from the October 2007 high and represents a decline of about 18% from October 2007. Please read my other posts http://9simplesteps.blogspot.com/2008/01/stock-market-major-declines-since-1940.html on the average number of times per decade that the market declines by 20% or more (answer: twice). So, if we get there, this would be historically typical and not an anomaly.

To get to a 20% decline, the DOW would have to hit about 11,300. That does not mean that we would stop at 20% if we got there since no one knows. As I often say anyone can tell you a hundred reasons why the market went up or down yesterday but that is not helpful and only makes for entertainment and hype.

If you have any desire to change your stock allocations, please consider that the recovery time after selling low is longer than retaining your positions. If you really are concerned (and I hope not) then please reassess your time horizon. If you need the money tomorrow it should not have been in stocks in the first place and if you can avoid touching the now lower values then you will shorten the inevitable time to recover.

Finally, only 4% of the time have stocks not made more, on average during a 10-year period, than inflation. We just experienced one of those 10-year periods (1999-2008). The next five years following such an occurrence have proved worthwhile for those who waited. http://9simplesteps.blogspot.com/2008/03/market-recovery-times.html

Will it happen this time? I have no idea. Maybe. Maybe not.

Market Recovery Times

As reported by Mark Hulbert, MarketWatch, March 7th, 2008 on market recovery times:


(1) The 2000-2002 bear market. The recovery time from this bear market took just four years, as judged by the dividend-adjusted version of the Dow Jones Wilshire 5000 index... (average that incorporates all publicly-traded U.S. stocks).



It was on Sept. 28, 2006, that the DJ Wilshire 5000 index surpassed its March 2000 high, almost exactly four years following the bear market low set on Oct. 9, 2002.


(2) The 1987 Crash. The recovery time from this crash, the worst in U.S. stock market history, took 17 months. It was on May 12, 1989, that the dividend-adjusted DJ Wilshire 5000 index surpassed its Aug. 25, 1987 high, less than a year and a half after the Dec. 4, 1987 low to which the stock market descended in the wake of the Crash.


(3) The 1973-1974 bear market. The recovery time from this bear market took just two years. By December 1976, the total-return version of the DJ Wilshire 5000 was higher than where it stood at the market's high in early 1973.


(4) The 1929 Crash and Great Depression. Because the DJ Wilshire 5000 index doesn't exist for those years, I turned to the stock series constructed by Jeremy Siegel, a finance professor at the Wharton School of the University of Pennsylvania, and author of the classic book Stocks for the Long Run. He shows that, for all intents and purposes, stocks on a total-return basis in late 1936 and early 1937 had risen back to their September 1929 high, before entering into another bear market. This puts the recovery time at a little more than four years from the stock market's July 1932 bottom.



Mark Hulbert suggests that it is the dividend-paying diversified index that makes the difference between a short recovery time and, for example, the QQQ (tech-oriented index) that still lingers 8 years later at 56% below its high. I could not agree more. Great article.

The P/E Ratio

The stock markets overall P/E ratio (Price of shares divided by earnings per share) averages 15-16 over the last 30-40 year period. Today the 12-month P/E forward looking is now just below 14 now that the stock market has declined about 17%. Some articles say that is cheap. Is it? Answer: "It depends". Real answer: "Who cares".

This period includes the highly inflationary late 70's and early 80's. When inflation was double-digit, the P/E on stocks was below 10 (in some years 8, yes, 8 in the 80's). If you remove those high inflation years, the P/E hovers more in the 18 range. That would lead me to believe that a P/E of 14 is about 20% (that is, 18-14 = 4; then 4 divide by 18) lower than the average. Maybe 30% if you use 20 as an average assuming we don't hit another period of double-digit inflation. If we do, then the current 14 is even too high, too, meaning stocks are still expensive.

So, yes, stocks may be cheap based on current inflationary trends. Maybe not. Even if inflation is 4-5%, that is way below the double-digit % that it was 25 years ago. Many, including me, believe inflation is much higher than the rates being stated these days (read my other labels on the economy-inflation for that) but, even so, they are not double-digit rates. But who knows where we are headed? I am not a market-timer. It does not work. If Nobel-winning economists cannot figure it out, then why do you think the local money manager (i.e. financial advisor, planner, consultant, whatever) can?

So the real answer is "Who cares". Why? Because you should not invest in the stock market thinking it is cheap now based on low P/E's or not. You invest because, long-term (10+ years), the stock market returns have a good chance of exceeding inflation and helping you meet your goals.

Stock Market Major Declines Since 1940

Here are some "rough estimates" of percentage declines and, more importantly, the amount of time to recover from past major market declines as measured by the DJIA - Dow Jones Industrial Average since 1940:


1946 (23%)...recovered in about 4 years

1957 (20%)...recovered in about 1 year

1962 (27%)...recovered in less than 2 years
1966 (25%)...recovered almost to previous high in 3 years, but then dropped again
1969 (36%)...recovered in about 4 years

1973 (45%)...recovered in almost 3 years, but then dropped again
1976 (27%)...recovered in 5 years, only to drop again

1981 (24%)...recovered in less than 2 years
1987 (36%)...recovered in less than 2 years (includes biggest one day drop of 22%)

1990 (21%)...recovered in less than a year
1998 (20%)...recovered in about a 1/2 year

2000 (37%)...recovered in 7 years, finally only to drop again
2008 (15% and still unknown past 1/22/08)...when will we recover to the 14,100 high of last year?

On average, since the 1960's, two declines per year of more than 20% with recovery times. The media will continue to report this news as if it is something unusual but it really is not. No fun as you go through it and wonder but markets do recover historically.

Fundamental Indexing

From one of my favorite professionals:

"... I think there's a lot of evidence that shows those very largest ones [are overvalued]. There's evidence that we've done on the S&P 500, where if you take the top market value stock every year over the 50 years of the existence of this index, there is tremendous underperformance of that strategy. Take the top 20 stocks, which is only 4 percent of the index in number, but one-third of the market value of the index, and it underperforms by more than 150 basis points per year over 50 years. So yes, those big-cap stocks do underperform and there is evidence that they do so significantly..."

Jeremy Seigel

A reason to consider Fundamental Indexing rather than Cap-weighted indexing.

NUA - if client has company stock

Is the cost basis important?

First, it's used to determine how much ordinary income tax a worker would pay if he transferred all or part of his stock into a taxable account.

Second, it's used to determine what the capital gains tax would be once the worker sells the stock.

For instance, if the cost basis is $10,000, the worker would be taxed in the year of distribution at the ordinary income tax rate on that amount. According to Keebler, any appreciation in the securities from the purchase date until the distribution date is known as NUA.
Thus, if a stock is worth $100,000 on the day the worker takes the stock out of the plan, the NUA would be $90,000. Better yet, the worker wouldn't have to pay tax until the stock is sold -- and even better yet, it would be taxed at the capital gains rate. Stock sold within 12 months could result in a combination of long- and short-term capital gains, while stock sold immediately after distribution or after 12 months will have just long-term capital gain. (Of note: workers who have any capital loss carryovers might want to take advantage of any NUA stock they own. Capital losses are fully deductible against capital gains so workers can use the NUA to wipe the slate clean, Cortazzo said.)

In some cases, workers might want to gift their NUA shares to a charity. That way the charity can sell without incurring any tax.
The other important aspect of NUA is this: To use NUA, the distribution must first qualify as a lump-sum distribution. According to Keebler, a lump-sum distribution is simply a distribution of all the assets in a qualified plan within one tax year. To get the special NUA tax treatment, however, one of the following triggering events must occur:
the distribution must be taken from a 401(k) or similar plan;
the entire balance must be paid to the account owner;
the entire distribution must take place within one tax year, by Dec. 31st.

To be sure, there's plenty to consider when it comes to NUA. Workers, for instance, who are under age 55 will have to pay a 10% penalty tax on the cost basis of the distribution. In addition, it's possible that some stock in the 401(k) plan qualifies for NUA and some doesn't, so be sure to separate those two types when transferring assets. According to Cortazzo, one difference between NUA and non-NUA shares is that NUA shares don't get a step-up in basis when inherited, while non-NUA shares do.
Lump-sum or rollover?
So what should workers do? Take the lump-sum distribution or the rollover?
According to Natalie Choate's bible on the subject, "Life and Death Planning for Retirement Benefits," most retiring employees should consider a rollover, the lone exception being when the lump-sum distribution includes appreciated employer stock. If you roll the stock into an IRA, the worker loses any chance to defer taxes on the NUA, plus rolling the stock into an IRA converts an unrealized capital gain into ordinary income.
According to Choate, factors to consider include the following:

How old is the worker? Young workers might want to consider the rollover while older workers might opt for the lump-sum distribution.

What other plans does the participant have? If the worker has plenty of money stashed in other retirement plans, the lump-sum distribution makes sense. But if the worker has just one plan, the rollover makes better sense.

How much of the distribution is NUA? If the NUA is a big portion of the plans' value, the NUA deal is more attractive. If the NUA is a small portion, the rollover is better.

As always, consult with a professional tax advisor or planner about the details.

Dividend ETF Index

Launched on June 16, WisdomTree Investments’ family of ETFs tracks WisdomTree’s proprietary indexes of dividend-paying companies. The fund family weights stocks in the index by dividend instead of the traditional method of market capitalization.

IBD explains that a company’s per-share dividend is multiplied by the number of outstanding shares. That number, the total cash paid by that company to its shareholders, is the company’s weight to the WisdomTree index.

It is an interesting concept. Remember that you need to be diversified because one segment of an asset class may do well for several years and then be taken over by another area (dividend-paying versus growth as an example) and no one knows when that will occur and for how long it will last.

These blog entries are not a recommendation for any particular financial asset or company but to inform you of what types of ooportunities and strategies exist in the marketplace. You must evaluate the fund prospectus thoroughly and understand what these products risks and objectives are and whether they fit into your asset allocation model. You do have an asset allocation model, right?