This post includes excerpts from the September 19th “Accounting for Inflation in Timber Forecasts” presentation at the 2013 “Who Will Own the Forest?” (WWOTF) conference in Portland, Oregon.
In 1913, one dollar bought 18 loaves of bread or three gallons of milk or close to three dozen eggs. Today, one hundred years later, one dollar fails to buy a single loaf of bread or gallon of milk. It buys about six eggs. This general increase in nominal prices, and associated decrease in real buying power, is inflation in action.
While inflation is a logical outcome of the money supply expanding faster than real growth, other factors apply. Research on average food prices over time by Jonathan Church and Ken Stewart at the U.S. Bureau of Labor Statistics (Beyond the Numbers: Prices & Spending, vol. 2, no. 6) demonstrates how consumer preferences and technology also influence prices. For example, the price of bread increased 25 times between 1913 and 2013. This tracks CPI – the Consumer Price Index – perfectly. Alternately, egg prices increased just 5 times, as advances in production, delivery, and storage techniques outpaced those seen for most other food items tracked by BLS. So on a relative basis, eggs got cheaper.
At times, when developing timber forecasts or reviewing timberland valuation models, we find that inflation is misunderstood. A single rising price of a single product or commodity is not inflation. Sawmills are making more money this year because lumber prices are higher than last year, not because of inflation, which tracks the overall price level. And yet, the real, relative costs and revenues affecting timberland investments and wood-using facilities do change over time at different rates and times. While cash flow estimates used to value potential investments include projections of future prices, we find honest confusion and common mistakes continue to plague the use of real and nominal rates and prices.
In developing timber forecasts in the U.S., we recommend quantifying historic relationships using nominal (stated) prices that include inflation, and then adjusting for inflation if necessary afterwards. Why? Because the basis of market clearing transactions and real-time negotiations between wood buyers and timber sellers at any point are the prices as known and stated at that time in each and every local timber market given the known and available supplies and technology. As timberland appraiser Jeff Wikle said at the WWOTF conference, “Yesterday’s (timber) price is based on yesterday’s technology.” Based on our understanding of forestry markets, this is exactly correct.
Once a price series has been forecasted, we can account for inflation (deflate prices), by dividing the series by a price index such as the Producer Price Index (PPI) or CPI. Applying the correct index is critical if we want to know the extent, in real terms, of trends or volatile price changes versus consistent increases over time. Which is better for deflating prices from timber forecasts: CPI or PPI? CPI highlights price changes over time for consumers, while PPI tracks changes in selling prices received by domestic firms providing goods and services. Typically, we use PPI, or a sub-series of PPI specific to relevant wood and forest products, because the growing and selling of timber or the manufacturing and selling of wood products represent, for the most part, business-to-business transactions.
Our research includes the evaluation of PPI tracking and alternate methods for estimating the effects of inflation over time for forestry-specific commodity prices. For example, price changes specific to primary wood inputs (logs, bolts, timber, pulpwood, and chips) indicate that log prices, on average, inflate at rates below PPI or CPI. Over the past fifteen years, while PPI and CPI have average annual inflation of just above or below 2.5%, wood raw materials averaged just below 1%. The data has limitations, but it provides a basis for thinking through how and why, in real terms, per unit timber prices may not accelerate. In practice, forest productivity continues to increase regionally, as does the efficiency of wood-using facilities. Carry this forward and we find a world that grows more wood faster, but needs less wood to produce the same volume of finished goods.
Does this guidance require investors to deflate nominal returns using PPI when evaluating their investment portfolios? No. The context of the analysis matters. In the case of investors, using CPI provides a logical and consistent approach to compare the value-to-the-investor of the real returns over time. However, when evaluating the costs and revenues “under the hood” of any given investment, we want to get as close as possible to the actual, real, relative values of these cash inflows and outflows over time to evaluate the critical value drivers. Otherwise, we fail to account for the relative role of technology, in particular, and supply-to-demand relationships over time.
To learn more about the Forisk Forecast or Forisk’s market-specific stumpage forecasts tailored to individual wood-using facilities or timberland ownerships, contact Brooks Mendell at bmendell@forisk.com, 770.725.8447.
To timberland appraiser Jeff Wikle’s observation at the WWOTF conference, “Yesterday’s (timber) price is based on yesterday’s technology” one should add “and yesterday’s market conditions.”
When using the PPI or the CPI, how do you account for the government’s frequent tinkering with how the indices are calculated? There is no good, single government calculated PPI or CPI that extends back 10, 20, 30, 40 or 50 years.
Robert Chambers
Big thoughts in a small package. Several are apparent:
1. Sawlog’s today are not of the same physical specs as those of yesterday. Currently sawlogs generally come from younger and thus smaller and more often plantation grown timber. Expecting the price per ton of today’s young, smaller and often lower grade sawlogs to return to the same price per ton of a decade or more ago may be wishful thinking. More narrow and lower grade lumber on average can only command so much from the market.
2. Current Sawmill technology often uses up to 40% less log volume to make the same lumber volume when compared to a decade ago…this is a staggering number. Other wood production processes have seen similar progress. A demand increase of 40% would be required to maintain the same demand for timber, and thus to presumably maintain price pressures.
3. On a Global basis, annual tree growth dwarfs tree removals from all sources. To shrug this off with comments like “most of excess timber is not economically available” is to deny the power of price to equalize supply. Tremendous quantities of timber and potential green field timber are currently not economically available due to lack of infrastructure including roads, rail, ports, mills and local economic demand. The acceptance of some form of capitalism by most governments and societies today can rather quickly improve infrastructure in response to price changes. New papermills in Uruguay providing markets for new plantations that did not exist a decade ago is a very strong example and there are others around the globe. (Note: A large papermill in Courtland, Alabama closes so the cosmos remains in balance, but I digress).
Too often, and I except it to happen again, wood product markets over-correct so that over supply soon dampens price enthusiasm and ultimately leads to price declines which leads to over-correction on the down trend and the cycle repeats. Unlike the hamster that always stays on the bottom striving for the top, we in the timber business often ride the wheel around the entire loop. And when we get back to the bottom or sadly from the top, (to paraphrase Robert Frost), “we like having said it so well we say it again, timber is a long term investment” and the wheel continues to turn.
North of the equator, we are just now beginning to tap the biological potential of our forest. The ambitious goal of sightly less than 2 green tons per acre from southern pine was set by a large integrated paper maker with their Dynamic Forest initiative of the 1960’s. By the mid to late 1980’s we were routinely expecting 4.0 or more green tons per acre and now we know that 10 or more tons per year is a realistic goal for the US South, not to mention even greater yields from all of the new Southern Hemisphere plantations.. As Bruce stated above “we find a world that grows more wood faster”. When one increases growth from 4 to 10 tons per acre each year then 60% fewer acres are required to produce the same volume. We may be planting fewer acres, but we are growing more wood per acre and probably more wood in total.
Less volume required per unit of demand in a world of excess timber (or potential green field timber) is not the recipe for continuous timber price increases. Keeping up with the Joneses may be easier than keeping up with inflation with Timber.
Any way, keep up the good work Bruce, an excellent and thought provoking article.