Forest-by-Forest Timber Sale Accounting for 1995

by Randal O'Toole


Results in Brief

Only five national forests made money on timber sales in 1995: the Inyo, Umpqua, Mississippi, Texas, and Allegheny. The Umpqua and Allegheny are perrenial money-makers, having valuable Douglas-fir and cherry respectively. The Mississippi and Texas are off-and-on money-makers, and this year they lucked out. The surprise is the Inyo, which usually loses money.

No surprise at all is the fact that all of the other forests lost money. While many in the Northwest, Southeast, and elsewhere could make money, the Forest Service has responded to declining timber sales by keeping larger shares of the receipts in the Knutson-Vandenberg, salvage sale, and other funds. In 1989, the agency kept about a third of timber receipts in these funds; in 1995 it was about two-thirds. Since counties get a quarter of nearly all receipts, that doesn't leave much left over for the Treasury.

This accounting of the number of forests making and losing money is based on the Thoreau Institute's simple criteria of comparing returns to the Treasury with costs to the Treasury. The Forest Service accounting system, known as the Timber Sale Program Information Reporting System (TSPIRS), uses the somewhat more liberal criteria of comparing all receipts, even those that don't go to the Treasury, with only some of the costs. But even TSPIRS concluded that 75 out of 120 forests lost money in 1995.

Every year for nearly a decade, the Forest Service has published its forest-by-forest accounting of the receipts and costs from national forest timber sales. The results generally come out in the May after the fiscal year in question.

The 1995 results were held up for nine months by the Clinton administration, which insists on micromanaging the agency even though it doesn't have time to do so (which is true of all micromanagers). Our analysis was held up even longer by the Forest Service's insistence that we file a freedom of information request for data not included in the TSPIRS files.

We hope to get 1996 results a lot quicker. But in the meantime, we are pleased to make the 1995 numbers available to anyone concerned about national forests.


What Does It All Mean?

Actually, TSPIRS is getting pretty meaningless after all these years. Everyone agrees that the Forest Service wastes a lot of money. After all, national forest timber sales have fallen by more than two-thirds since 1990, yet national forest timber budgets have fallen by just a third.

TSPIRS admits that most forests lose money. In 1995, TSPIRS reported losses on 75 forests out of 120. Total losses on those forests were more than $80 million while the net for both profitable and unprofitable forests was minus $57 million.

TSPIRS greatly underestimates the actual losses to taxpayers. The Thoreau Institute's reanalysis says that in reality 116 forests lost money, and losses from those forests totalled nearly $375 million.

But once you admit there are losses the actual numbers are mere quibbles. The question that is still being debated is: Do these losses represent subsidies to the timber industry? In large part, the answer is "no." In fact, most of the losses represent subsidies to the Forest Service bureaucracy.

The timber industry may get two kinds of subsidies from below-cost timber sales.

So in general, the losses in the sale program more reflect a bloated bureaucracy than corporate welfare. The timber sale losses are symptoms of serious problems with the Forest Service. Demonizing the industry, however, is probably not a constructive response.

General Methodology

TSPIRS is actually three different reports:
  1. The financial account, which supposedly represents the cash flow produced by timber sales;
  2. The economic account, which supposedly represents the returns on investments in national forest timber; and
  3. The income and employment account, which supposedly describes the number of jobs and personal earnings due to national forest timber.
In fact, all three of these accounts contain serious flaws.
  1. The financial account doesn't display the cash flowing in and out of the Treasury because it mixes dollars paid by the Treasury with dollars paid out of timber receipts.
  2. The economic account doesn't accurately show returns on investments because it mixes receipts today with investments in future timber growth.
  3. The employment account doesn't accurately show jobs from national forest timber sales because it includes inappropriate multipliers for secondary employment.
Most of the Thoreau Institute's reanalysis is devoted to turning the financial account into a true cash-flow analysis. The financial account reports all timber receipts and all timber costs. However, it amortizes many of the costs using an inappropriate formula. This leads to an underestimate of cost and, since many of the receipts don't make it to the Treasury, an overestimate of receipts.

The Institute's reanalysis requires four steps:

  1. Identifying the unamortized costs, all of which are displayed on a computer printout titled "TPIR01." Receipts are also on this form.
  2. Distinguishing the costs paid by the Treasury from the costs paid out of the KV, salvage sale, or other Forest Service fund. The TPIR01s distinquish KV, brush disposal, and road maintenance funds, but not salvage sale funds. So the Forest Service provided the Institute with data showing how much each national forest spent out of the salvage sale fund on each line item in TSPIRS.
  3. Distinguishing the receipts paid into the Treasury from the receipts paid into the KV, salvage sale, or another Forest Service fund. TSPIRS never distinguishes this information. So the the Forest Service provided the Institute with data showing how much of each national forests' receipts went into the KV and other funds.
  4. Allocate the regional and Washington office (including experiment station) timber program costs to the individual forests. Although TSPIRS counts these costs, it doesn't count them against receipts. The Institute divides these costs up according to the volume of timber sold by each forest.
One slight problem is that the Forest Service's supplemental salvage sale data combines sale preparation costs with the costs of doing NEPA analyses. These two costs are distinguished in TSPIRS. The Institute subtracted the salvage sale cost from the TSPIRS sale preparation cost, which means that the posted sale preparation costs will be underestimated (and occasionally negative), while the NEPA costs will be overestimated.

The Data Files

The Institute has posted three tab-delimited data files on the web. We have made no attempt to make the columns line up, so you should download the files to your computer and open them with a spreadsheet. The files are:
  1. The key TSPIRS data file includes the basic results of TSPIRS and the Institute's reanalysis of Forest Service accounting data;
  2. The detailed TSPIRS data file includes line-item-by-line-item accounts of timber sale costs as taken from TSPIRS documents;
  3. The supplemental data file accounts for all data, including a breakdown of receipts by fund and expenses paid out of the salvage sale fund, not included in original TSPIRS documents.
In order to properly attribute receipts and costs to the same timber, TSPIRS is an accrual accounting system that attaches costs and receipts to the actual volumes of timber cut. If no timber is cut on a forest in a given year, TSPIRS reports no receipts or costs even if the Forest Service actually spent or received money that year for timber cut in other years. All of the data in all three files are based on this same accounting stance.

The accrual accounting system is fine for most expenses. But TSPIRS is deceptive in two basic ways:

Correcting these two problems produces an accurate comparison of the amount of money the U.S. Treasury receives and spends on timber sales in each national forest. Thus, the Institute's accounting stance is that of taxpayers who want to know what they are getting for their money.

Rows in All Three Data Files

To make sure that the rows in all three data files line up, the files have some blank spaces. The following general pattern is used, though some files do not have all the rows. All of the columns in the above three files are described below.

Key TSPIRS Data

The key TSPIRS data file consists of nineteen columns of data as follows. Many of the columns in the key file are combinations of columns in the detailed and supplemental files. At the risk of confusing people, each of these combinations is listed below. Some rows are intentionally left blank so that the file lines up with the detailed data file.
  1. Forest name.
  2. Total timber revenues, including purchaser road credits and funds kept by the Forest Service.
  3. Expenses as calculated by TSPIRS. A detailed breakdown of these expenses is in the detailed TSPIRS data file. As noted above, this number is generally an underestimate of actual timber sale costs.
  4. TSPIRS net shows column 2 minus column 3. This is the Forest Service's claimed return on national forest timber sales.
  5. Revenues kept by the Forest Service includes timber receipts going into the Knutson-Vandenberg, salvage sale, brush disposal, road maintenance, and other funds. In 1995 the Forest Service kept almost half of all timber receipts.
  6. Road credits represent the in-kind payments recorded when purchasers built roads and credited the costs of those roads against the amount they bid for timber.
  7. County payments represent funds paid to the counties in lieu of property taxes. On most forests, this is slightly less than 25 percent of the total, but on the "spotted owl" forests it is often much more. This is because Congress gifted those counties with a declining floor on payments when timber sales declined to protect the spotted owl.
  8. Kept by the Treasury shows the funds left over after deducting the Forest Service funds, road credits, and county payments from total receipts. On many forests, the Treasury actually lost money before counting any other costs.
  9. Sale costs are the TSPIRS costs actually paid by taxpayers on such activities as sale preparation and administration. Costs paid out of the K-V, salvage sale, and other timber receipts funds are not included. This column is equal to columns 10, 12, and 29 from the detailed date file minus columns 9 through 13 of the supplemental data file. This is the cost that will be slightly underestimated because the supplemental salvage sale data do not distinguish NEPA costs.
  10. Road costs are costs paid by taxpayers on roads, mostly for engineering and design. This column is equal to columns 21 and 28 from the detailed data file minus column 16 from the supplemental data file.
  11. Reforestation and timber stand improvement costs are costs paid by taxpayers for these activities over and above the KV reforestation costs. This equals columns 30, 31, and 33 of the detailed data file.
  12. Forest overhead represents timber's share of a forest's general administration costs. This equals columns 11 and 41 of the detailed data file minus column 17 from the supplemental data file.
  13. Regional and Washington office overhead shows costs of the timber program spent by other offices and allocated to the forests based on the volume of timber they sold. Regional costs are divided among the forests in the region weighted by volume cut, while Washington costs (including the forest products lab through the Intermountain station) are divided among all forests in the country weighted by volume cut.
  14. NEPA, appeals, and litigation shows the costs of writing environmental documents and dealing with challenges to timber sales (many brought by timber companies questioning the terms of sale contracts). This equals columns 13, 14, and 36 from the detailed file minus columns 14 and 15 from the supplemental file. This cost is slightly overestimated because the supplemental salvage sale data do not distinguish NEPA costs.
  15. Total costs is the sum of the previous six columns.
  16. Net to Treasury shows the receipts kept by the Treasury (column 8) minus the costs to the Treasury (column 15).
  17. The volume cut is shown in million board feet.
  18. Acres cut include cutting by all methods, ranging from selection cutting to clearcutting.
  19. Jobs include both primary and secondary jobs attributable to timber. In reality, the secondary jobs are pretty speculative. To estimate only the primary jobs--loggers, truckers, and millworkers--divide by three.

Detailed TSPIRS Data

The detailed data file includes 42 columns of data. Except for column 42, all of these numbers are from the column titled "DATA FROM TPIR MST" in the Forest Service report TPIR01. These numbers are essentially unmassaged by the Forest Service.

Before TSPIRS calculates total expenses (column 3 in the key data file), the road and reforestation costs are heavily discounted. Column 42 hints at the extent of this discounting.

  1. Forest name.
  2. Receipts for timber.
  3. Timber road credits.
  4. Other timber charges.
  5. Interest & penalties.
  6. Total receipts, the sum of the previous four columns.
  7. Harvest administration costs for commodity sales.
  8. Harvest administration costs for stewardship sales.
  9. Harvest administration costs for personal use sales.
  10. Total harvest administration costs, the sum of the previous three columns.
  11. General administration costs associated with timber sales.
  12. Sale preparation costs.
  13. Costs of environmental analyses and documentation.
  14. Costs of timber sale appeals and litigation.
  15. Brush disposal costs (paid out of the brush disposal fund, not tax dollars).
  16. Road maintenance costs (paid out of the road maintenance fund, not tax dollars).
  17. Engineering and construction costs associated with road surfacing.
  18. Engineering and construction costs associated with road culverts.
  19. Engineering and construction costs associated with bridges.
  20. Engineering and construction costs associated with road prisms (the part of the road below the surface; often the most expensive part of construction due to necessary earth movements).
  21. Total road engineering and contruction costs.
  22. Purchaser road engineering and construction costs associated with surfacing.
  23. Purchaser road engineering and construction costs associated with culverts.
  24. Purchaser road engineering and construction costs associated with bridges.
  25. Purchaser road engineering and construction costs associated with prisms (the part of the road below the surface; often the most expensive part of construction due to necessary earth movements).
  26. Total purchaser road engineering and contruction costs.
  27. Transportation planning.
  28. Silvicultural examination costs.
  29. Genetic tree improvement costs.
  30. Appropriated reforestation costs.
  31. Reforestation costs paid by the Knutson-Vandenberg fund.
  32. Appropriated timber stand improvement costs.
  33. Timber stand improvement costs paid by the Knutson-Vandenberg fund.
  34. Facilities depreciation costs.
  35. Other appeals and litigation costs (primarily associated with roads).
  36. Timber program general administration (as distinct from timber sale general administration).
  37. Total expenses.
  38. Total receipts minus costs.
  39. Volume cut in thousands of board feet.
  40. Total of capitalized road assets. This represents the sum of all road costs that have not been counted against timber because of the Forest Service's bizarre amortization formula.

Supplemental Data File

The supplemental data includes two kinds of information:
  1. The first seven columns (after the name of the forest) break down the receipts by fund, including the Knutson-Vandenberg, salvage sale, and other funds kept by the Forest Service. This information is not provided by TSPIRS.
  2. The next sixteen columns show how salvage sale funds were spent by each forest. TSPIRS does not discriminate between spending out of tax dollars and spending out of the salvage sale fund. To correctly show the effect of sales on the Treasury, spending out of the salvage sale fund must be deducted from TSPIRS expenses.
The twenty-four columns in this table present the following data:
  1. Forest name.
  2. KV receipts shows how much of the timber receipts went into the Knutson-Vandenberg fund. This and the next four columns add up to column 5 in the key TSPIRS file.
  3. SSF receipts shows how much of the timber receipts went into the salvage sale fund.
  4. BD receipts shows how much of the timber receipts went into the brush disposal fund.
  5. Road Maintenance receipts shows how much of the timber receipts went into the road maintenance fund.
  6. Other Forest Service receipts shows how much of the timber receipts went into other funds kept by the Forest Service.
  7. NFF is the National Forest Fund, the account into which national forest receipts are paid that should be retained by the Treasury. Negative numbers indicate that the Forest Service transferred dollars paid into this fund in previous years to another fund such as KV.
  8. KV OH is based on 1996 numbers but we include it here because it is so intriguing. It shows the percentage of KV receipts that timber sale planners set aside for administrative overhead. If the percent is 50, for example, then half of all 1996 KV collections are set aside for overhead. On most forests it is about a third, but at least one forest set aside more than two-thirds of KV funds for overhead. On some forests, the amounts vary by ranger district; the numbers shown are the unweighted averages of all districts.
  9. Timber inventory (silvicultural examination) costs paid by the salvage sale fund.
  10. Timber resource planning costs paid by the salvage sale fund.
  11. Timber support costs paid by the salvage sale fund.
  12. Sale preparation costs paid by the salvage sale fund.
  13. Harvest administration costs paid by the salvage sale fund.
  14. Appeals & litigation costs paid by the salvage sale fund.
  15. The costs of reworking plans after appeals & litigation paid by the salvage sale fund.
  16. Timber road costs, primarily engineering and design, paid by the salvage sale fund.
  17. Sale overhead (general administration and program management) costs paid by the salvage sale fund.
  18. Ecosystem management costs paid by the salvage sale fund.
  19. Lands activities (probably land line location) costs paid by the salvage sale fund.
  20. Facilities costs paid by the salvage sale fund.
  21. Law enforcement costs paid by the salvage sale fund.
  22. Ecosystem management overhead costs paid by the salvage sale fund.
  23. Total costs paid by the salvage sale fund.
  24. Timber sale costs paid by the salvage sale fund. This is the sum of columns 9 through 17, and these costs are deducted from the appropriate columns in the detailed TSPIRS data file to get the sale, road, and other costs in the key TSPIRS data file. Columns 18 through 22 are not carried over to the key data file.

TSPIRS' Bizarre Amortization Formula

The Institute's accounting of national forest timber receipts and costs differs from TSPIRS in two main ways:
  1. The Institute counts only receipts that stay in the Treasury, while TSPIRS counts all receipts including in-kind payments and receipts retained by the Forest Service.
  2. The Institute counts costs as they are paid by the Treasury, while TSPIRS uses a faulty amortization formula for roads and reforestation.
The TSPIRS amortization formula supposedly "pools" road and reforestation costs and counts only the share of the pool attributable to timber cut this year. The size of the pool is based on the volume of timber that the Forest Service expects to cut during one rotation.

A "rotation" is the number of years the Forest Service expects to grow trees before cutting them down. This is typically about 50 to 120 years.

Say that a particular national forest spends $1 million per year on roads and reforestation. Each year, $1 million is added to the pool. Then an amount equal to the percentage of a rotation's worth of timber that is cut this year is deducted from the pool. That is the amount that is counted as costs this year.

If a forest's rotation is about 100 years, then an average of 1 percent of a rotation's timber is cut each year.

As it turns out, even though the actual cost remains $1 million per year, as the pool fills up the cost counted by TSPIRS approaches $1 million asymptotically but never reaches it. After 1,000 years it is still $43 short of $1 million; after 1,800 years it is still $1 short of $1 million. So the pool system guarantees that TSPIRS will always underestimate actual costs.

The reality is even worse than this hypothetical example. First, TSPIRS uses no adjustment factors for inflation. Road costs of twenty or thirty years ago, which were far lower than today because of the inflation since then, are added to the pool as if there were no inflation. This underestimates those costs in the same way that simply adding ten British pounds to ten dollars and calling it twenty dollars would underestimate the value of the pound.

Second, TSPIRS relies on FORPLAN estimates of the volume of timber to be cut in a rotation. But more than half of the national forest FORPLAN models projected far higher timber cuts after twenty to fifty years than in the first ten years of the plan. The extra cutting in later years was from acres included in the suitable timber base that were not needed to sustain the first decade's level of cut. Thus, the volume cut in the first year is a disproportionately small percentage of the rotation.

For example, say a forest with a 100-year rotation planned to cut 20 billion board feet over a rotation but only 100 million board feet per year during the first ten years. Then the share of the rotation cut each year is only 1/2 percent, not 1 percent, so TSPIRS will count only 1/2 percent of the pool cost against this year's timber receipts. Yet when cuts go up in later decades, twice as much annual road contruction will be needed to access the additional timber.

The third problem is that the actual cost will not remain $1 million per year--it will probably grow much larger. Once built, roads do not last forever: They require reconstruction every twenty years or so. This means that, at the end of a 100-year rotation, five miles of road will require annual reconstruction for every mile of road built per year in the first 100 years. The annual costs of reconstruction will probably be much higher than the initial annual construction costs.

For all these reasons, the Forest Service's amoritzation formula makes no sense. Rather than try to figure out a "correct" formula, the Thoreau Institute simply counts road construction and reforestation as an operating cost. In the big picture of an entire national forest over the long run, this is much more accurate that trying to capitalize these costs.


Electronic Drummer