Accounting for fixed assets places a significant burden on the corporate tax function. Tax departments invest a tremendous number of staff hours tracking fixed assets and grappling with constantly shifting rules and regulations. This already difficult situation was compounded by the release of the tangible property regulations effective for tax years beginning in 2014 and after.
A statistical sampling analysis can ease the process by reducing the number of fixed assets that need to be analyzed in detail. A sample of a few hundred assets may be used to estimate tax values for thousands of assets. The IRS not only accepts statistical sampling but encourages it in evaluating tangible property. (See Rev. Proc. 2011-42 for the statistical methodology taxpayers may use to estimate values for federal tax returns. In addition, tangible property revenue procedures that cite sampling solutions include Rev. Procs. 2011-43, 2012-19, 2012-20, 2014-16, 2014-17, 2014-54, and 2015-20.)
In a tangible property regulations study, a number of accounting methods and elections may be considered, some of which require filing Form 3115, Application for Change in Accounting Method. Commonly, these include changes in depreciation recovery periods and repairs and maintenance items deductible for tax purposes. Further, the tangible property regulations allow a taxpayer to elect to recognize the remaining basis upon a partial disposition of an asset, as well as for the corresponding removal cost—which can be deducted in the year of disposition. An efficient study may include all four.
In a tangible property regulations study using sampling, the scope of the study is established by targeting the assets with potential benefit. For efficiency, these assets are grouped into strata, or "buckets," and sampled separately. Tax determinations are made on only the sample items, and these results are used to extrapolate the total benefit for all of the targeted assets, not just the sampled ones.
Incorporating Estimated Values Into the Fixed-Asset Listing
Once the statistical sample yields estimated results, the next step is to incorporate those estimated values into the existing fixed-asset listing. Early planning can streamline and facilitate implementation into the fixed-asset software package. This item discusses common industry solutions for updating the fixed-asset listing after sampling and best practices to consider when planning a tangible property regulations sample.
End-Stage Updating the Fixed-Asset Listing: The Basics
While it is somewhat circular, envisioning the fixed-asset updates at the end guides the planning and designing at the beginning—long before a single sampling unit is ever pulled. This item examines two basic approaches. By this point in a study, the sample has been pulled, analysis of the sampled records has been completed, and total amounts to recategorize have been extrapolated. Now what?
Most approaches require driving down those overall total estimated extrapolated amounts to the individual asset level. That is, every asset in the targeted scope gets an allocated amount of the total estimate. For example, a sample may estimate that 80% of all capitalized roof improvements should be recharacterized as repairs and maintenance expenses. Then, for each roof asset in scope, 80% of its cost is assigned to repairs and 20% is capitalized in its original life. From this point, while there are a wide variety of approaches, the more common industry practices represent many variations on two themes: either "shrinking" or "exploding" these estimated asset-level allocations to update the fixed-asset listing with the sample results.
Shrinking: Some taxpayers opt to shrink the estimated asset-level allocations by summing up the estimated amounts by placed-in-service year and life—thus updating the fixed-asset listing with a handful of adjustments for each year in scope.
Example 1: Suppose an estimated $100 million in cost basis should be moved from 39-year modified accelerated cost recovery system (MACRS) assets to five-year MACRS personal property. Of this, $40 million is in the current year, $35 million is in the prior year, and $5 million is in the year before that, and so on. Some taxpayers will create two new assets entries for each placed-in-service year. That is, in the current year, one asset is negative $40 million with a 39-year recovery period, and the other is positive $40 million with a five-year recovery period. In the prior year, there is a negative $35 million 39-year asset and a positive $35 million five-year asset, and so on.
Pros: This is simple and makes it easy to update the fixed-asset listing. Only a few additions to the fixed-asset listing result from the study. It is not necessary to completely remove the fixed-asset listing and upload a new file.
Cons: Future changes to the fixed-asset listing are more difficult. Specific identification of disposals becomes a best-guess exercise. Further, partial dispositions of a structural item are allowed under the tangible property regulations and can be a significant opportunity each tax year. The IRS requires the taxpayer to be "reasonable" when ascertaining the value of the partial disposal. Shrinking results makes it difficult to reach a reasonable value. Also, tracking placed-in-service dates becomes cumbersome, which can lead to inappropriate depreciation calculations. The shrinking method is not a recommended best practice.
Exploding: By contrast, other taxpayers explode the estimated asset-level changes by creating a new asset for each asset in scope.
Example 2: Suppose 5,000 assets were capitalized as 39-year MACRS property before the study. As a result of the sample, it is estimated that 30% of the original cost basis in 39-year MACRS property should be moved to five-year MACRS property, 10% should go to 10-year MACRS property, another 10% should go to 15-year MACRS property, and 50% should remain as 39-year MACRS property. The 5,000 assets are removed, split into pieces, and replaced with 20,000 new assets. That is, an asset that was originally $100,000 in the 39-year recovery period is replaced with four new assets based on the extrapolated percentages; one is $30,000 with a five-year recovery period; another is $10,000 with a 10-year recovery period; the third is $10,000 with a 15-year recovery period; and one is $50,000, remaining in the 39-year recovery period. All four new assets carry the same placed-in-service date as the original asset.
Pros: It is easier to maintain this fixed-asset listing going forward. Identifying partial and full dispositions becomes a reasonable exercise. The tangible property regulations improvement rules rely on building systems and building components to make facts-and-circumstances decisions of whether to capitalize or expense an item. A sound fixed-asset foundation with proper detail allows these decisions to be made, which often leads to tax savings. Also, placed-in-service dates of each asset maintain their integrity, allowing for proper depreciation.
Cons: This produces a much larger fixed-asset listing with substantially more changes. However, software-savvy tax consultants can facilitate an automated upload. Further, they can set up rules (discussed below) that will greatly ease the level of explosion needed to best represent the results and produce logical records for tax purposes.
Variations on these themes mostly involve how the asset-level estimates are made before shrinking or exploding. These include:
1. Change the sample records to the values as found when reviewed, and only extrapolate the asset-level allocations to nonsampled assets. This is slightly more complex but has fewer records than in the exploding approach because the sampled assets are not exploded this way.
2. If sampling was performed at a project or location level, rather than asset by asset, first extrapolate to the project or location level, then allocate these estimates down to the asset level. When extrapolating to the asset level, allocate different percentages depending on the stratum where the asset was placed for sample selection. Often, for sampling purposes, assets may be grouped according to anticipated reallocation percentages. Group assignment may be based upon keywords found in the asset description, such as "HVAC" or "tiling." It may be based on size as measured by the original cost basis. When there are large, medium, and small designations, typically, a higher proportion of repairs are found among the smaller assets.
3. Similar to the variation above, assign different percentages based on attributes of the asset even if the sample was not stratified by these attributes before it was selected. This is more work for the statistician, but it is a feasible alternative.
Taking into account practical considerations and using a little finesse during the sample design and planning stages can ease the implementation and facilitate future use and maintenance of the fixed-asset listing.
Planning With an Eye on the Target
Effective teamwork between the tax department and statisticians will make a strategic difference and result in a much more successful study. During the scoping stage, assets are identified for potential opportunity—whether that be recovery period, repairs and maintenance, partial disposition, or removal cost changes. But the tax department should not stop there and just hand over the targeted assets for the statisticians to sample.
Possible outcomes: Possible outcomes for an asset need to be understood from the end-stagefixed-asset-implementation perspective so that the eventual statistical sampling estimates will be incorporated into an updated fixed-asset listing in a way that logically makes sense and facilitates future use of the fixed-asset listing.
Example 3: An asset labeled "water piping loop" with an acquired value of $50,000 is identified under a keyword of "piping" for a potential recovery period change and/or a repairs and maintenance change. By the placed-in-service date, if that water piping loop is a component of the building's original construction, a rule can be created that this asset should not receive any extrapolated repairs and maintenance allocation, as the original asset should not be recharacterized as a repair. By contrast, suppose the water piping loop was capitalized in the current year as an improvement to a building that was placed into service 10 years prior. Another rule can be created so that these kinds of assets do not receive any extrapolated partial disposition allocations.
It is important to note that the allocation rules are to be followed no matter whether the asset is actually selected for the sample or has extrapolated values from the sample results applied to it.
Guiding the sample design: The scoping and planning at the beginning of a study is the ideal time to determine how the desired fixed-asset updates will be framed and what rules may be created to ease the eventual end-stagefixed-asset implementation.
Allocation rules can be used in determining whether strata are warranted in the sample design. For example, the $50,000 water piping loop may be found to be part of an expansion to the manufacturing plant that occurred five years after the initial construction of the plant. Since the expansion is considered original construction, the rule would be that no repairs and maintenance should be extrapolated to it, only recovery period changes.
The tax team and statistician may then decide to bucket original construction projects into their own stratum—to separate them from the post-construction assets, which are the only ones that should receive any repairs and maintenance allocations. This stratification allows for an easily implemented estimation approach to achieve the required allocation rules.
Alternatively, the same end may be achieved in a simpler sample design with less stratification by using more complex estimation approaches. The deciding factor of which alternative to use may be the cost. Sometimes stratification will add to a study's costs, and sometimes it will not; in fact, strategic stratification often can improve precision of estimated values from smaller samples. Still, in very small studies, the extra bucketing may not be practical; statisticians then will have to perform more estimation work to implement the rules.
While the exact solution may vary, the important factor is to put the potential outcomes on the table for consideration when planning a study.
Bonus depreciation: Bonus depreciation is another consideration. Tax law has varied over the years, and bonus depreciation should be applied only to the eligible assets in the eligible time periods. During the planning and designing stages, assets eligible for bonus depreciation can be identified, and the team may consider stratifying based on bonus eligibility, or alternatively preparing for the estimation work around when the sample results are allocated.
Building componentization: The tangible property regulations require a taxpayer to compare an asset to one of eight building systemswhen deciding to capitalize the asset as an improvement or to expense it as a repairs and maintenance expense. In addition, the tangible property regulations allow a taxpayer to elect a partial disposition. Both situations require taxpayers to think about building components and how they are treated in the fixed-asset system. Therefore, in the same vein, the implementation should consider components—not only to best reflect the analysis but to allow easier deduction of further repairs and maintenance and partial dispositions in the future.
Conclusion
While updating the fixed-asset listing may be the last step in implementing a tangible property regulations study, CPAs should carefully consider prior to the analysis what the implementation will look like. There are many solutions, based on the facts and needs of the taxpayer, and each requires varying amounts of effort to execute. Identifying these needs during the planning stages can influence decisions in the sample design, so for a sample design best suited to facilitate implementation in this final step, the CPA should be sure to communicate these needs and work with the tax team and statistician.
By being thoughtful about statistical sampling implementation in the planning stages, the CPA can help improve the framework of the sample design, which will, in turn, increase the focus and efficiency of the engagement and allow for an easier and more sensible execution.
The sampling results may be implemented in various customized ways into the fixed assets, and in today's tangible property regulations environment, taxpayers are looking for assets to be split into component detail. The planning for this final step should start during project inception, as it will directly affect the level of effort and time needed to provide the desired detail at the end of the study.
EditorNotes
Greg Fairbanks is a tax managing director with Grant Thornton LLP in Washington.
For additional information about these items, contact Mr. Fairbanks at 202-521-1503 or greg.fairbanks@us.gt.com.
Unless otherwise noted, contributors are members of or associated with Grant Thornton LLP.
