If a Tree Falls in a Forest, Does It Generate an Adequate Return?

By Daniel Rohr, CFA | 09-11-09 | 06:00 AM

This is the first of two articles on timberland investments, one area of the broader forest product industry. In Part I, we will introduce readers to the asset class, its risk and return characteristics, and recent investment trends. We'll also take a brief look at publicly traded timberland owners Weyerhaeuser (WY), Plum Creek Timber (PCL), and Rayonier (RYN), which collectively own 16 million acres of United States timberland.

Timberland is an unusual asset. In some ways, it's a lot like a copper deposit or oil reservoir, since resource extraction can be deferred, allowing inventories to be "warehoused" when pricing is soft without risk of obsolescence or spoilage.

In other ways, timberland assets share more in common with corn or sugar cane fields. If managed properly, investors can expect a steady stream of cash flows into perpetuity, whereas a copper deposit or oil reservoir ceases to generate income when the ore runs out or the well runs dry.

Like the corn or sugar cane field, timberland investments also include an option value embodied in the land beneath the trees (or crops, as the case may be). If the land's value for "higher and better use" (e.g., property development) exceeds the net present value of future logging cash flows, the investor is best served by selling the land for alternative use.

Another way to look at these two components of timberland value creation is to think of the asset as we might a convertible bond. Timberland possesses the continuous (albeit variable) cash flows of a bond, but also the capital appreciation upside of a stock. Regardless of how you prefer to look at it, timberland's unusual qualities make for an interesting set of risk and return characteristics.

Timberland Risk and Return
Before we take a look at timberland from a risk and return perspective, we should caution that there's no perfect data set with which to do so. Timberland isn't traded in deep, liquid markets like stocks and bonds, and no parcel of timberland is exactly the same. Price discovery, therefore, is neither instantaneous nor all that reliable. Initial attempts to develop timberland performance data took place in university economics, agriculture, and forestry departments, where researchers constructed a variety of synthetic return series based on log prices in a particular region. More recently, through the efforts of institutional investors and academics, we've seen the emergence of timber indexes, notably the NCREIF Timberland Total Return Index and the now-defunct Timberland Performance Index.

The NCREIF return series, initiated in 1987, has since become the accepted benchmark for the asset class. The NCREIF Timberland Total Return Index, as the name suggests, includes an income component and a capital appreciation component. The total return data, which are publicly available on NCREIF's Web site (the income and capital appreciation components are not), are presented in the following table.

A quick glance at the annual returns should quicken the heart of any profit-seeking investor. A hypothetical investor who put $10,000 into timberland in 1988 would have ended up with $158,096 by the end of 2007, assuming he or she could duplicate the NCREIF results. Before taxes and fees, an investor in the S&P 500 would have seen his or her initial $10,000 pile grow to $93,341. And that's before the dark days of late 2008. On a risk-adjusted basis, timberland looks even better, boasting a set of risk and return characteristics that would make Bernie Madoff blush. The following table presents a standard set of performance metrics for timberland and an investor's typical toolkit of asset classes covering the 20-year period of 1988 through 2007. Worth noting is that, for reasons intrinsic to the NCREIF data (timberland appraisals are typically conducted annually), we used the annual return series for each data set--not ideal, but a sacrifice we must make for the purposes of undertaking this comparison.

Institutions Follow the Money
Not surprisingly, timberland's attractive combination of strong returns, low variability of returns, and relatively strong correlation with inflation caught the attention of large institutional investors, ranging from groundbreaking Yale endowment manager David Swensen to the country's largest pension plan, CalPERS. Following the lead of trailblazers like Swensen, total institutional timberland investments increased from about $1 billion in 1989 to $8 billion in 1999 and about $14 billion in 2002. Latest estimates put the cumulative number close to $40 billion.

Where does timberland fit into institutional portfolios? Most institutions include it in their real asset allocation, a broad category that includes assets ranging from real estate to commodities like oil and gold. Harvard's endowment includes a 9% target allocation to timberland (up from 2% in 1998). CalPERS allocates about 34% of its inflation-linked asset class to timberland as well as 10%-20% of its real estate portfolio, a substantially larger share than it allocated to timberland a decade ago.

Such is their desire to obtain timberland exposure that institutional investors like Harvard and CalPERS have dominated the buy side of major timberland deals over the past several years. According to the timberland transaction database maintained by Forestweb, timberland investment management organizations--the appointed agents of pension plans and endowments--have accounted for all but one of the 14 purchases greater than 500,000 acres since 2003 (a total of about 15 million acres).

So who's doing the selling? Primarily, the forest product industry: paper and packaging companies like International Paper (IP) (5.6 million-plus acres sold for $6.6 billion in 2006) and MeadWestvaco (MWV) (323,000 acres sold for $400 million in 2007) and lumber manufacturers like Temple-Inland (TIN) (1.6 million acres sold for $2.4 billion in 2007). According to the U.S. Forest Service, by 2006, nearly 80% of forest product industry land had been snapped up by timberland investment management organizations and real estate investment trusts, marking nothing less than a complete overhaul of the once fully integrated forest product industry's supply chain. Today, just a handful of publicly traded forest product companies still have substantial timberland investments. Among the top 10 largest nongovernment timberland owners, only three are not timberland investment management organizations: Weyerhaeuser, the only classic forest product company on the list, and timber real estate investment trusts Plum Creek Timber and Rayonier.

It's fairly clear from the preceding table that the market assigns substantial value to timberland holdings. If it didn't, Plum Creek would trade a much lower multiple, closer to that of Weyerhaeuser or Rayonier, which derive a much smaller portion of their earnings from timberland. Given what we know about timberland's risk and return profile over the past 20 years, perhaps it's not surprising that the market assigns Plum Creek this kind of premium multiple.

But should it? How should we value timberland and the companies that own it? We need to consider what the next 20 years might look like for timberland. With the transition from industrial to investor ownership of timberlands now nearly complete, is the capital appreciation of the past two decades a one-time, fund-flow-driven occurrence? Or do timberland values have more room to run? On the cash-flow front, what does the uncertain outlook for housing starts mean for lumber-related income? Will the secular decline in paper demand hurt timberland owners? And what about the prospect of deriving energy from wood in the form of power generation or biofuels? Fundamentally, we'll need to address questions of supply and demand, both for the products made from wood as well as for timberland itself. In Part II of our look at timberlands, we'll attempt to do so.