Thursday, June 30, 2005

Philosophy of Wooden Deck Management

The retiring principal at our local elementary school shared a bit of wisdom with the graduating fifth graders last week: What you think determines what you say, which should be consistent with what you do. In this post, I am going to tell you how I am thinking about the wooden deck in my backyard and what I am doing to solve a problem with it. This discussion is an example of the type of long-term mindset that helps with investing too.

We inherited a huge one-thousand-square-foot deck from the previous owner of our house, who at some point in the deck's history decided to paint it solid beige, presumably to match the color of the siding of the house. Over the years the latex paint has started to chip and peel in places and boards have begun to rot. Wood experts tell me that deck boards painted on just the top surface do not hold their paint for long, because water gets into the wood from the sides and bottom, and a sauna-like effect from the sun beating down on wet wood capped on top with a layer of paint literally pushes the paint up and off. The result is the unsightly chipping and peeling that we see.

Through talking to deck professionals and people at paint stores, I have learned that paint works well on siding (which is typically less exposed to direct sunlight and rain), but oil-based stain (not latex paint!) is the customary finishing for decks. Stain preserves properly prepared wood by penetrating deep into its pores (instead of sitting like a gummy rubber glove on its surface as latex paint does) and wears off gradually rather than chipping and peeling, i.e., stain ages more elegantly than paint. From a maintenance point of view, there is also a significant difference between paint and stain--paint on a deck lasts maybe five years or so before it starts to crack and chip off, requiring considerable effort at that point to remove (by loosening, scraping, sanding, etc.) prior to repainting; whereas stain needs to be reapplied about every couple of years, and it is a lot easier to strip off old stain residue (more of a washing than a scraping process) in preparation for restaining.

Because I have an interest in seeing my existing wooden deck "live" as long as possible, I have embarked on the one-time (I believe) task of removing all of the paint and then properly staining the deck. I am finding that I must rely on a combination of tools and tricks of the trade to remove the paint (failing in places but still stubbornly adhesive in others): "friendly" chemical strippers (biodegradable, with sodium hydroxide (lye) as the caustic agent), water-misting of the surface to keep the chemical stripping process active, careful scraping in the direction of the grain of the wood with a non-metal edge after the paint has softened (more vigorous scraping is possible with the scraper perpendicular to the wood), and water-washing with the nozzle of a garden hose (pressure washing at 1000 psi or higher would damage the denuded wood surface). Importantly, also essential are many hours of unwavering perserverence and physical labor, and plenty of patience and restraint (not to wash off the stripper too soon, nor to scrape too hard, nor to zap with a power washer, nor to stop even though my joints and muscles ache from too much crouching and stooping).

When I eventually reach my objective of removing all of the paint (well, let's say 99% of it, as I hear that it is virtually impossible to remove all of the paint), I will have gotten set up for proper deck maintenance the way that I would like to have it--gradual and evolutionary (occasional cleaning and restaining), instead of drastic and repeatedly revolutionary (tedious and catastrophic paint removal prior to repainting each time). As I strip off the latex paint, I am beginning to catch glimpses of an ancillary aesthetic reward--the natural beauty of the flowing grain and subtle color variations of the wood, previously hidden behind a veil of opaque beige paint. My long-term thinking allows me to justify dedicating a month of my attention now to the deck, to reposition myself on the right track towards deck longevity and greater visual appeal.

Back to the principal's parting words: Following his thinking-speaking-doing message, he concluded by mentioning how one's actions ultimately reveal one's character as a person and participant in our society--probably about as deep a message as eleven-year-olds are able to grasp (especially after having sat attentively and listened for the previous half hour). In a similar vein, my decisions and actions taken during my summer deck project tie in with my values as a person--managing for the long haul, utilizing existing tools and materials to their fullest, treading lightly on the environment, having the wherewithal to make drastic changes when needed to get back on the right track, and exhibiting a lot of self-reliance. In my opinion, these traits--surfacing here in deck management--are useful ones to have in long-term investing as well.

Sunday, June 26, 2005


The most important part of investing is "finishing the race," by which I mean creating an investment portfolio that is at least self-sustaining and hopefully even grows substantially in value over time.

Recently I have been reading Collapse: How Societies Choose to Fail or Succeed (2004), an informative and insightful book by Jared Diamond about past and present societies and the factors that determine their survival or demise. The author's point is that societies both historically and today have a choice that greatly influences their future. For example, the former inhabitants of Easter Island in the South Pacific, originally settled around 900 A.D., perished largely because they cut down too many old-growth trees, many of which they used to transport and erect hundreds of massive stone statues (later apparently toppled in internal conflict between rival clans) around the perimeter of the island (these stone statues appear on lists as one of the new Seven Wonders of the World). With the disappearance of the largest trees, away went the shade vital to proliferation of lower-lying plants and wildlife, gradually turning self-sustaining island paradise into a barren environmental desert. By the mid-1700s, following a lifespan of around 800 years, the original society on Easter Island had collapsed.

At the other extreme, Professor Diamond mentions Japan, which (as I learned to my surprise) today has the highest percentage of forested land (74%, versus some 20% for the U.S.) among all First World countries in the world today. This impressive statistic is partially the result of Japan's moist climate and mountainous terrain, but is also the outcome of a timely shift in governmental policy at a critical juncture. By the 17th century, Japan's old-growth forests had largely been harvested, with trees having been used in general construction, for rebuilding following large-scale fires in Edo (Tokyo), to erect castles as ornamental symbols of the wealth and power of regional lords (analogous to the stone statues on Easter Island), to build ships to wage war (unsuccessfully) against Korea, and for various industrial purposes (e.g., smelting iron, firing tiles and ceramics, and making salt). The disappearance of the forests led to flooding, river siltation and other problems. Fortunately, the Tokugawa government realized that this new environmental crisis had been created by human-driven deforestation, and proceeded to implement an elaborate long-term reforestation and woodland management program. Today, most of the timberland in Japan sits in managed forests stemming from this government-led tree planting endeavor begun some 300 years ago.

An investment portfolio is similar to a society, in that it either flourishes or depletes over time, depending on many factors, both external and internal. The external factors are the economy, government policy, corporate behavior (leadership, honesty, competition and alliances), taxation, etc. The important internal factors are the financial decisions that we as managers of our own porfolios make regarding level of risk, buy-sell timing, diversification, turnover, payment of fees to outside managers, etc. Our own investment management decisions impact results very significantly and are the critical element over which we can exercise direct control.

In long-term investing, building a sustainable portfolio is the way to make sure that one at least finishes the race. A sustainable portfolio is one that produces income to cover any cash flow needs and also appreciates in value to give a total rate of return in excess of inflation. Just as every society has a choice that affects its longevity and ultimate survival, each of us as investors has management choices that influence our portfolio performance and financial future. Despite all of the perils in the modern investment world, the good news is that controlling one individual's behavior during a lifetime (i.e., your own, as manager of your own portfolio during your own lifetime) ought to be so much simpler than changing an entire society's cultural habits and shifting its course over many centuries!

Friday, June 17, 2005

Should a Millionaire Be a Coupon Clipper?

I sometimes hear that how carefully a person watches his expenses is a sign of whether or not he has "made it," indicating that wealthier people spend their time and energy maximizing their earnings, assets and net worth rather than concerning themselves with "petty" day-to-day expenses. Under this view, any millionaire who takes the time to clip a coupon from a local supermarket ad to save a dollar on a gallon of milk is guilty of "thinking small," since he really should be focussing on how to realize an extra one percent (over the course of a year, that's $10,000 for a millionaire!) on his investment portfolio. In other words, why waste time clipping coupons when there are more important financial decisions to make? At least, that's how the conventional argument goes.

My own view differs from the conventional view presented above, since I organize my thinking around control and predictability of outcome rather than ranking the importance of decisions by their dollar amounts. I illustrate by selecting examples from different areas of personal financial management:

A. Expense: Applying for a credit card offering a 1% to 2% cash rebate on all purchases. Potential gain: $300 per year. Certainty: 100%. Expected gain: $300;

B. Income: Cancelling vacation plans at boss's request because an important project has come up at work. Potential gain: $3,000 increased bonus. Certainty: 50%. Expected gain: $1,500;

C. Investment: Shifting funds into a more risky investment with a higher expected return. Potential gain: $30,000 in one year. Certainty: 10%. Expected gain: $3,000.

Applying conventional logic to these three examples, one ought to be inclined to pursue B and C above more diligently than A, since their expected payoffs are higher, i.e., their outcomes involve larger amounts of money.

I arrive at the opposite conclusion, however, by engaging a "control factor" in my decision-making process. In line with the old proverb about how "a bird in the hand is better than two in the bush," I focus on what I can control (the bird in my hand) rather than devote time to endeavors beyond my control (the two birds in the bush). This control factor is what makes applying for the rebate credit card (in example A) a "no brainer" decision for me (I know that I will get a $300 rebate), while making the investment decision (in example C) a less obvious one (I dislike the potential downside that is beyond my control). Outcomes that I can control with very little effort are the "low hanging fruit" of personal financial management that I believe it makes sense to harvest, before venturing off into areas with less controllable outcomes.

Based on this "control factor" rationale, I find myself clipping milk coupons. I know with very high certainty that I can save a dollar by taking a minute to grab the coupon before heading out the door to the store. Regardless of how high one's net worth is, I contend that there will typically be seemingly "petty" decisions involving only very small amounts of money that clearly make sense to pursue. Could this be why Benjamin Graham's advice was to pick stocks the way you shop for groceries, or why Warren Buffett is known for picking up pennies he finds on the floor of the elevator?

Thursday, June 09, 2005

What Causes Residential Real Estate to Appreciate?

Before reading on, please take a moment to answer the above question by selecting the best answer from the choices below:

A. High household income
B. High population
C. Rising household income
D. Rising population

We all know that real estate prices are higher in urban than in rural areas, and highest of all in the affluent residential enclaves of our largest cities. But, how did these high prices get to where they are? In other words, what are the essential factors that cause real estate prices to rise?

Thinking about human settlement patterns, urbanization and industrialization throughout the course of civilization, my initial guess was that rising household income (choice C) and rising population (choice D) would be the most important drivers of residential real estate prices. I figured that, not income levels and populations themselves but instead changes in these variables would be most highly correlated to rising real estate prices. After all, classical equilibrium economics tells us that, not supply and demand themselves but changes in supply and demand are what drive changes in prices.

To explore this issue, I use data from the U.S. Census Bureau ( for median household income and population and from the Office of Federal Housing Enterprise Oversight ( for changes in median housing prices. Interested in long-term price behavior, I go as far back as the OFHEO data allow (to 1980), capturing changes over the past 25 years. To achieve a large enough but still manageable sample size, I focus on the 50 states plus the District of Columbia, giving 51 distinct geographical entities in total.

Below I summarize the data, showing extremes, averages and correlations:

(To help display correlations, I provide in parentheses beside each data point the corresponding annual percentage change in median housing prices for the period 1980-2005)

Median Household Income in 1984 (A)
1. Alaska: $32.3k (3.4%)
2. Connecticut: $30.0k (5.9%)
3. Maryland: $29.7k (5.9%)
4. Hawaii: $28.9k (5.9%)
5. New Jersey: $27.8k (6.5%)
47. Alabama: $17.3k (3.7%)
48. West Virginia: $16.8k (3.1%)
49. Tennessee: $16.8k (4.0%)
50. Arkansas: $15.7k (3.4%)
51. Mississippi: $15.4k (3.2%)
Average: $22.4k (5.1%)

Population in 1980 (B)
1. California: 23.7 mil. (6.9%)
2. New York: 17.6 mil. (7.2%)
3. Texas: 14.2 mil. (2.7%)
4. Pennsylvania: 11.9 mil. (5.1%)
5. Illinois: 11.4 mil. (4.9%)
47. District of Columbia: 0.6 mil. (6.8%)
48. Delaware: 0.6 mil. (5.9%)
49. Vermont: 0.5 mil. (5.6%)
50. Wyoming: 0.5 mil. (3.1%)
51. Alaska: 0.4 mil. (3.4%)
Average: 4.4 mil. (5.1%)

Annual Percentage Increase in Median Household Income from 1984 to 2003 (C)
1. Tennessee: 4.3% (4.0%)
2. Wisconsin: 4.3% (4.5%)
3. District of Columbia: 4.3% (6.8%)
4. Minnesota: 4.1% (5.1%)
5. Alabama: 4.1% (3.7%)
47. Montana: 3.0% (4.5%)
48. Texas: 2.9% (2.7%)
49. New Mexico: 2.8% (3.9%)
50. Oklahoma: 2.8% (2.4%)
51. Alaska: 2.5% (3.4%)
Average: 3.5% (5.1%)

Annual Percentage Increase in Population from 1980 to 2002 (D)
1. Nevada: 4.6% (5.1%)
2. Arizona: 3.2% (4.5%)
3. Florida: 2.5% (5.3%)
4. Alaska: 2.2% (3.4%)
5. Utah: 2.1% (4.2%)
47. Pennsylvania: 0.2% (5.1%)
48. Iowo: 0.0% (3.4%)
49. North Dakota: -0.1% (3.1%)
50. West Virginia: -0.4% (3.1%)
51. District of Columbia: -0.5% (6.8%)
Average: 1.1% (5.1%)

Annual Percentage Increase in Median Housing Price from 1980 to 2005
1. Massachusetts: 8.0%
2. New York: 7.2%
3. Rhode Island: 7.0%
4. California: 6.9%
5. District of Columbia: 6.8%
47. Wyoming: 3.1%
48. North Dakota: 3.1%
49. Louisiana: 2.9%
50. Texas: 2.7%
51. Oklahoma: 2.4%
Average: 5.1%

Correlations Between Increase in Median Housing Price and:
A. Initial Median Household Income: Correl. = 0.45
B. Initial Population: Correl. = 0.08
C. Increase in Median Household Income: Correl. = 0.15
D. Increase in Population: Correl. = 0.07

The data reveal that, moreso than changes in income and population (C and D, low correlations), it is median household income itself (A, correl. of 0.45) that appears to be the most significant driver of housing prices. This is an unexpected result, indicating that the mechanism underlying appreciation of residential real estate prices over at least the past 25 years has NOT been the traditional model of people flocking from countryside to city for work, receiving higher pay and concurrently bidding up real estate values in the cities through increased demand and limited supply. Based on percentage increase in median household income, Tennessee and Wisconsin rank #1 and #2; yet, when ranked by real estate appreciation, they fall to #35 and #26, respectively. Similarly, population has risen most dramatically in Nevada and Arizona; however, in terms of real estate price appreciation, these states rank #19 and #25, respectively.

Another way to view the data is to see how the top 15 states by housing price appreciation compare with their corresponding rankings across our four factors:

Ranking by Housing Price Appreciation vs. Ranking by Factors: A, B, C, D

1. Massachusetts: 6, 12, 36, 37
2. New York: 25, 3, 27, 43
3. Rhode Island: 26, 41, 14, 35
4. California: 12, 2, 23, 9
5. District of Columbia: 38, 48, 3, 51
6. New Jersey: 5, 10, 19, 26
7. Maine: 35, 39, 41, 32
8. New Hampshire: 8, 43, 6, 14
9. Connecticut: 2, 26, 39, 39
10. Delaware: 9, 49, 34, 16
11. Hawaii: 4, 40, 43, 20
12. Maryland: 3, 19, 46, 19
13: Vermont: 23, 50, 31, 24
14. Virginia: 7, 15, 15, 15
15. Washington: 13, 21, 33, 10

Average Rankings: 14 (A), 28 (B), 27 (C), 26 (D)

Among our four factors, only initial median household income (A) carries a significantly high average ranking, being in the teens instead of the mid-twenties--hence, exhibiting a much higher correlation to price appreciation than the other factors do.

How, then, do we explain our result that housing prices have appreciated most significantly in states where median household income started off 25 years ago being higher than average? I offer the following plausible explanation:

1. Maturing Society: During the past 25 years, we have seen a maturing of American society, with established cities and neighborhoods receiving more attention than newer communities. Driving influences include the aging of the babyboomer generation and their lifestyle preferences (social "herding"), growth of the service sector and office-related work (favoring in-city workplaces), etc.

2. Wealth Concentration: The result is increased demand for real estate in already established areas where incomes are already high and cultural amenities are more developed. Existing income drives real estate prices higher, producing additional wealth for existing property owners and further concentrating wealth in these areas.

3. Attraction to the Coast: Among the top 15 states showing the greatest real estate price appreciation, all are located in the Northeast or Mid-Atlantic (between Maine and Virginia) or on the West Coast (including Hawaii), and only Vermont has no coastline exposure on the Atlantic or Pacific Ocean. Geography (ocean, beaches), climate (milder temperatures) and business and trade patterns (with Europe and Asia) all contribute to the attraction of people and businesses to coastal states and cities.

Based on this trend of coastal property price appreciation, I would venture to guess that, for better or worse, over the next 25 years we will see further polarization of American society into older, more established "haves" residing in affluent communities on the Atlantic and Pacific Coasts and contributing to a further rise in real estate prices in the coastal states, versus a larger group of "have nots" living in more inland areas with a lower wealth base and experiencing less property price appreciation.

In essence, real estate price appreciation trends are an example of how wealth (high household income--choice A) tends to beget more wealth (manifested, in this case, by rising real estate prices). Surprisingly, the absolute level of income and wealth (as opposed to changes in income levels) appears to have the most significant impact on real estate appreciation.

Wednesday, June 01, 2005

Herding, Helixes and a Hunch About Higher Prices

Humans are by nature social creatures. In both stock and real estate investing, this social "herding" behavior can be used to explain much of the price trending that we see, with markets often continuing to rise (or fall) for longer than would be consistent with a random walk model of price movement. In the stock market, we saw the impact of herding during both the spectacular dot-com frenzy in 1999-2000 (i.e., on the way up) and the equally spectacular bursting of the tech bubble in 2000-2001 (i.e., on the way down). In the real estate market, as some market commentators warn, recent double-digit year-on-year percentage price appreciation could prove to be too speculative for sustainability.

Just as "prudent" investment advice cautions investors against overweighting high P/E Internet stocks, real estate pundits warn homebuyers to think twice before buying real estate on either the East Coast or the West Coast of the U.S., pointing out how affordability is better in the Midwest and South where year-on-year percentage price appreciation remains in the single-digit range. In essence, whether the subject is stocks or real estate, here we have further examples of the familiar debate between value and growth, mean reversion and trending, Old Economy and New Economy, or status quo and change.

If history were to repeat itself in lock-step fashion, simply retracing its path as time and events cycle along, those in the value camp with an investment philosophy based on mean reversion would consistently outperform growth investors betting on change in the New Economy. However, since the value vs. growth debate continues to rage on, it should be clear that prices move neither in predictable cycles (mean reversion) nor in a straight line up or down (trending). A simple model that captures both types of movement is a forever-ascending helix. Civilization and markets do cycle around (i.e., history repeats itself) but at the same time humans do learn and markets do tend to advance in secular fashion (i.e., we make progress and markets rise).

Based on this helical model of human achievement and market prices, I have a hunch (i.e., it's difficult to "prove" but I intend to provide some "evidence" in later posts) that over the long run a) Internet stocks will run much higher than Old Economy stocks, and b) coastal real estate will outperform inland markets. There will, of course, be times when markets overshoot on the upside and downside (the cyclical part of the helix) due to short-term herding behavior among investors; yet, interestingly enough, on a longer time scale this social herding instinct among humans is the very element that can predictably drive seemingly overpriced markets still higher (the trending part of the helix).

As the 21st century unfolds, I envision an increasingly Net-centric future for our society with lifestyle choices and climate preferences driving profits for both a) companies successful in addressing the social needs of us all to stay "connected" to others and b) real property owners in the coastal markets experiencing population growth stronger than in inland markets. Similar to the impressive growth in population and property values in some of the world's largest cultural and financial centers over the centuries--Shanghai, Bombay, Seoul, Tokyo, New York, London, Hong Kong--I expect continuing above-average growth in large U.S. cities on the Pacific and Atlantic coasts--San Francisco/Bay Area, Los Angeles, Seattle, Portland, New York, Boston, Washington, D.C.

The investment implication here is a simple prescription: Long term, invest where the herd is heading (which is not necessarily the same as where the herd is investing)--towards a Net-centric society, with population concentrated on the West and East Coasts. Though prices of Internet stocks and coastal real estate may seem high today, the fundamentals (i.e., Internet usage and the influences of climate and geography) are, in my opinion, robust enough to continue to drive demand and prices much higher still.