3. Measuring and Reporting Corporate Income Tax Expenditures

Non-standard deductions, exempt and non-taxable income

The OECD recommends that revenue forgone from base narrowing TEs (i.e. non-standard deductions, exempt and non-taxable income) is measured using the CIT rate that each company is subject to. That is, (i) the standard CIT rate, (ii) the reduced rate if the company benefits from a special regime or (iii) a higher rate for companies in the financial sector. Such an approach is preferred over DIAN’s current approach, which multiplies the tax base narrowing amount that is claimed with the standard CIT rate regardless of whether the company pays the standard rate or not. Even though the standard CIT rate is the benchmark rate, multiplying the base narrowing amount with the standard rate biases the revenue foregone estimate upwards (in the case of reduced rates) and provides a wrong indication of the actual revenue foregone that can be attributed to the base narrowing provision. In fact, the DIAN approach combines the effect of two TEs: the base narrowing provision and the impact of the reduced CIT rate.4

Revenue forgone across the total number of firms $n$ from a provision $p$ that grants exempt income ${Exempt income}_{p, i}$ is equal to:

${Revenue\mathrm{ }forgone}_p={\sum }_{i=1}^n\left({Exempt\mathrm{ }income}_{p,\mathrm{ }\mathrm{ }i}\right)\mathrm{*}{Rate}_i$

Where ${Rate}_i$ stands for the CIT rate business $i$ is subject to. The same formula applies to non-standard deductions or special deductions and non-taxable income.5

The revenue forgone from enhanced deductions is equal to the additional deduction exceeding the standard deduction that is included in the benchmark, multiplied by the company’s CIT rate. If the benchmark stipulates full deductibility of the costs incurred (e.g. for the cost linked to salary payments), the additional deduction would be calculated by multiplying the deductible item (e.g. the salary payment to first-job employees under the age of 28) with the difference between the allowed deduction under the enhanced deduction (e.g. 120%) and the standard benchmark deduction (e.g. 100%). Revenue forgone from a provision $p$ that grants an enhanced deduction for amount $Enhanced\mathrm{ }deductio{n}_{p,\mathrm{ }i}$ then equals:

${Revenue\mathrm{ }forgone}_p={\sum }_{i=1}^n\left(Enhanced\mathrm{ }deductio{n}_{p,\mathrm{ }i}\right)\mathrm{*}{({DD}_{p,\mathrm{ }enhanced}-{DD}_{p,\mathrm{ }benchmark})\mathrm{*}Rate}_i$

Where ${DD}_{p,\mathrm{ }enhanced}$ stands for the currently allowed deduction (in percentage) and ${DD}_{p,\mathrm{ }benchmark}$ for the permitted deduction under the benchmark (in percentage).6

Reduced rates

Revenue forgone from reduced rates can be split into two separate components:

1. 1. The first component multiplies the difference between the standard rate and the reduced rate with taxable income, in line with the approach that DIAN currently follows. In this calculation, taxable income corresponds to the actual taxable income of each business that has been reduced by all other base narrowing TEs that the business can claim (e.g. deductions, exemptions, or non-taxable income). The first category therefore attempts to isolate the revenue forgone from the reduced rate by abstracting from the interaction between the reduced rate and other TEs.

   The following formula illustrates the proposal in the case of FTZ businesses; the same formula applies to other special regimes (mega-investments, stability contracts, hotels, etc.). The revenue forgone from the reduced rate ${Rate}_i$ of 20% applicable to $k$ FTZ businesses with current ${Taxable income }_i$ would be equal to:

${Revenue\mathrm{ }forgone}_{FTZ}={\sum }_{i=1}^k\left({Taxable\mathrm{ }income}_i\right)\mathrm{*}(Rat{e}_{benchmark}-{Rate}_i)$

1. 2. The second component reflects the interaction between reduced rates and the TEs that narrow the tax base (i.e., non-standard deductions, exempt income, and non-taxable income). This amount is equal to the revenue forgone that could (potentially) be recovered if Colombia would decide to abolish both types of TEs (i.e. the reduced rate as well as the base narrowing TEs). The revenue forgone from the interaction is obtained by multiplying the total value of the special deductions, exempt income and non-taxable income for each firm $i\mathrm{ }$with the difference between the standard CIT rate and the reduced rate, and then summing the result for all companies subject to the reduced rate.7

   The revenue forgone from the interaction for $k$ FTZ businesses subject to a reduced rate ${Rate}_i$ of 20% would be equal to:

${Revenue\mathrm{ }forgone}_{Interaction,\mathrm{ }\mathrm{ }FTZ}={\sum }_{i=1}^k{\left(\sum Deductions+\mathrm{ }\sum Exempt\mathrm{ }income+\sum Non-taxable\mathrm{ }income+\sum Enhanced\mathrm{ }deduction\mathrm{*}({DD}_{\mathrm{ }enhanced}-{DD}_{benchmark})\right)}_i\mathrm{*}(Rat{e}_{benchmark}-{Rate}_i)$

The presentation of the TEs into these 2 components would allow DIAN to (i) isolate the revenue forgone from different type of base-narrowing TEs and the reduced CIT rate; and (ii) demonstrate how the different types of TEs reinforce each other and thereby result in additional revenue forgone. Nonetheless, it should be noted that breaking down the revenue forgone in such a way is merely a matter of displaying total revenue forgone in a more transparent manner.

Overall, this break down of total revenue foregone into three separate components (i.e., the TE corresponding to tax base narrowing discussed in the previous subsection, the reduced tax rate and their interaction discussed in this subsection) would increase transparency; the sum of these three components would be equal to the difference between (benchmark tax base x benchmark tax rate) and (current tax base x current tax rate).

Deferrals

While accelerated tax depreciation allowances do not alter the total amount of the depreciation allowances that can be deducted over the life of an economic asset, they do change the moment in time, and therefore the tax value, when these depreciation allowances can be claimed. Measured on a cash-flow basis, the revenue foregone of accelerated tax depreciation in a given year equals the revenue forgone due to the additional depreciation allowances that can be deducted in the year, relative to the amount that would be deductible under the tax benchmark, which are tax depreciation allowances that match the economic depreciation of the asset.8

Revenue forgone calculated on a cash flow basis is defined by the cost of a TE to the government in a given year.9 Accelerated tax depreciation results in positive tax revenue forgone initially, followed by periods during which revenue receipts are higher than they would have been under tax depreciation that equals the economic deprecation of the asset (because a higher percentage of the asset’s value has already been deducted). Ultimately, the total depreciated value is equal to the cost of the asset. Under the proposed TE benchmark, tax depreciation allowances match the economic depreciation of the asset over the asset’s useful economic life. Because of the impact on tax revenue in different time periods, the cash flow-based estimate of accelerated tax depreciation is not only the result of the accelerated depreciation of investments made in the current year, but the revenue forgone in any given year is also impacted by the accelerated deductions claimed with respect to investments dating from previous time periods (Department of Finance Canada, 2012[4]). For that reason, the net cash-flow fiscal cost in a given year could be positive or negative depending on past, current and projected investments (Department of Finance Canada, 2021[5]; Villela, Lemgruber and Jorratt, 2010[6]). Due to this period-specific impact of tax deferrals, detailed information about investment patterns over time would be required to correctly determine the size of the TE.

The cash flow revenue forgone estimated in the current period could be defined as follows:

${Revenue forgone}_{p, t}=\sum _{i=1}^n(Effective depreciatio{n}_{p,t,i}-Economic depreciatio{n}_{p,t,i})*Rat{e}_{t,i}$

Where $Effective depreciatio{n}_{p,t,i}$stands for the allowed amount of depreciation for all assets that fall under the accelerated depreciation provision $p$that were depreciated by business $i$in year $t$. Some of these assets may have been purchased in $t$ while other may have been purchased in previous years. The economic depreciation term refers to the value that would have been depreciated for asset $p$ in firm $i$ if the depreciation schedule had followed the economic depreciation rates (i.e. the benchmark).

The Canadian TE report (Department of Finance Canada, 2012[4]) identifies five key methodological steps to measure revenue forgone from accelerated tax depreciation:

1. 1. Determine the useful life10 for each type of investment subject to accelerated depreciation. If the asset cannot be matched to a particular asset type, use a weighted average of the most similar asset types. For example, clean energy generation equipment may include a broad variety of assets with potentially different benchmark depreciation rates.

2. 2. Create, for every taxpayer, a separate time series on an asset-by-asset basis including all assets that benefit from accelerated tax depreciation as reported in the CIT return.

3. 3. Recalculate claims and balances under the useful life (i.e. economic) depreciation rates (counterfactual).

4. 4. Estimate the total current-year revenue forgone by comparing actual tax depreciation in the year to the depreciation allowances under the TE benchmark (i.e. the counterfactual).

5. 5. Break down the total revenue forgone by separating the impact of current and prior year’s investments.

Tax credits

The revenue forgone from tax credits is equal to the total amount claimed by companies that either (i) reduces the final tax liability or (ii) is refunded. Not all tax credits are viewed as a TE – for example, the foreign tax credit for taxes paid abroad by resident taxpayers earning foreign source income is not viewed as a TE under the proposed benchmark. DIAN is therefore advised to list revenue forgone from tax credits on an item-by-item basis in addition to providing an estimate of total revenue forgone.

Revenue forgone estimates

This section provides item-by-item revenue forgone estimates for Colombia in 2020. The OECD produced preliminary estimates based on a representative sample of CIT tax returns that was merged with exogenous information. The coding has been shared with DIAN so that the proposed methodology could be applied to all taxpayers. This report includes the results of the calculations made by DIAN, following the OECD’s recommendations, and based on the methodology described in the previous section.

The figures included in Table A and Table B were calculated by DIAN based on exogenous information that was merged to tax returns. As only 44% of corporations in Colombia are required to file these supplementary forms, the estimates represent a lower bound of the revenue forgone associated with a specific TE.

The totals in the tables also represent a lower bound estimate because a large fraction of TEs cannot be measured on an item-by-item basis due to a lack of data. For this reason, the revenue forgone linked to the “other” category is likely large. For example, revenue forgone from other exempt income (including some non-TE components such as exempt income from Andean Community tax treaties) was equal to COP thousand million 3 961 in 2020. If Colombia were to follow the recommendations made in section 3.5, the design of future tax returns would ensure that TEs and tax provisions that are part of the TE benchmark are not included together in a single box. Consequently, the table is not comparable to figures reported in the MFMP (2022). A tentative full list of CIT TEs is included in Annex C.

Reporting recommendations

This section also provides recommendations on how to report CIT revenue forgone in the TE report:

• Whenever possible, DIAN should report revenue forgone instead of the aggregate value of tax exemptions, non-taxable income, deductions, or tax credits. Because businesses can pay a CIT rate that is higher or lower than the standard rate (due to surtaxes or special regimes), revenue forgone cannot be easily inferred from the aggregate size of TEs.

• The main CIT revenue forgone table (Table A) lists subtotals by type of TE (tax credits, deductions, etc.) and, where data is available, includes rows with item-by-item estimates.

• For each type of TE, the table also contains a row that records the revenue forgone that cannot be measured on an item-by-item basis (“Other”). Currently, no estimate can be provided in these rows because the aggregate boxes in the tax return (for example, total exempt income) include both TE and non-TE components.

• Over time, DIAN could add more information in the table by adding the revenue forgone in previous years. This will indicate how the revenue foregone has changed over time.

• In the TE report, values would be expressed both in thousand million pesos and as a percentage of GDP.

• A total of all measured CIT TEs could be included; this total should however be interpreted with caution given the difficulties of summing different TEs. If the total is provided, the report should explicitly mention that the revenue forgone concept does not take into account the interaction between the CIT TEs, and that some TEs are not included in the total because of lack of data. In line with the recommendations made in the preceding section, rows that report revenue forgone from the TEs that cannot be disaggregated within a category (Other), would have to be calculated using the company-specific CIT rate rather than the standard CIT rate.

• In addition, DIAN could include an additional table (Table B) that reports total revenue forgone from TEs for firms that are subject to special regimes. Even if all revenue forgone linked to companies under special regimes was included in Table A, Table B would summarize all the revenue forgone that is measured only considering companies that benefit from special regimes. Hence, it would be possible to identify revenue forgone from base narrowing TEs aside from the interaction component. Table B would therefore complement the data in Table A.

• The 2022 edition of the MFMP includes a column that records the number of firms that benefit from a reduced rate and are subject to each special regime (FTZ, stability contracts, etc.), in line with the recommendation of the Tax Incentives Commission (OECD, DIAN and Minhacienda, 2021[3]). In the future, the number of firms subject to each special regime could be included in Table B.

Item-by-item discussion

In addition to the tables that provide an overview of revenue forgone from individual CIT TEs (Table A) and revenue forgone linked to companies subject to special regimes (Table B), the TE report could include item-specific tables for each TE measured on an item-by-item basis. These tables would explain why a specific provision is considered a TE, provide a brief description of the data source and estimation method used to calculate the revenue forgone from that particular TE as well as include information about the reliability of the revenue forgone estimate. The table would indicate if the estimate is based on tax return data or exogenous information.

The item-by-item discussion would also provide additional context to guide the interpretation of the revenue forgone figure. For example, when discussing the revenue forgone of a particular TE, it should be mentioned whether the TE corrects for a distortion that is created by another TE. For example, the discussion of revenue forgone from the CIT credit granted for investment in fixed assets should refer to the fact that, in contrast to the proposed benchmark, businesses in Colombia cannot recover the VAT paid on investment (negative VAT TE). Table 3.3 provides an illustrative example.

More specifically, the item-by-item table should include the following elements:11

• (i) TE identifier; (ii) Name of the TE; (iii) Revenue forgone in the current year and in the five preceding years; (iv) Tax type; (v) TE type; (vi) Year of implementation; (vii) Year of expiration; (viii) Legal reference; (ix) Reliability of the estimate; (x) Data source used to estimate revenue forgone; (xi) Estimation method; (xii) Paragraph that states why the provision deviates from the benchmark and provides more detail about the TE.

Full list of TEs in appendix tables

Finally, a comprehensive list of TEs such as the one included in the appendix of this report (Annex C) could be included in the appendix to the TE report.12 Ideally, this list should provide information about the legal reference and a brief explanation of the extent to which the tax provision deviates from the benchmark. This list would include all TEs, and not just those that were measured on an item-by-item basis.