Investors require financial data that is comparable over time, comparable within a single set of financial statements, and comparable between companies. Unfortunately, this is not always the case. We explain how differences between IFRS and US GAAP, accounting policy options, differing interpretations and accounting estimates, can all reduce comparability.
Convergence and comparability should be a priority for the IASB and FASB. Present consultations by the IASB and FASB regarding the accounting for credit losses are a good opportunity to better align IFRS and US GAAP, and to remove the confusing disconnect between purchased and originated loans, as we discuss in our response.
In our recent article ‘Expected credit losses: Beware the day 2 effect’ we explain how the accounting for loan loss provisions leads to a lack of comparability (and we argue a lack of relevance) of the financial statements of banks. Comparability is compromised by the different approach to the accounting for originated loans compared with (some) purchased loans. However, the biggest issue for investors is the difference between the accounting under IFRS and US GAAP, with the latter producing a higher provision and an earlier expense in profit and loss.
Following the 2008 Global Financial Crisis, both the FASB and IASB developed a more forward looking ‘expected credit loss’ model for loan impairments. Initially the two boards worked together to produce a globally converged approach; however, for various reasons, including managing the conflicting views of different prudential regulators, a common standard never materialised. Unfortunately for investors, the present requirements differ significantly.
FASB and IASB consultations on loan impairment accounting make no mention of convergence
Both the FASB and IASB are now consulting investors and other stakeholders about their respective loan impairment accounting standards. FASB has issued an exposure draft of a proposal to expand the use of their gross-up method (the main subject of our earlier article), while the IASB has issued a more general consultation as part of their post implementation review of IFRS 9. Neither consultation mentions convergence. It is an ideal opportunity for the two boards to once more work together to try and finally produce a converged approach to loan losses and improve comparability for investors in this area.
We have written a letter addressed to both FASB and IASB in response to their respective consultations in which we urge renewed efforts to find a common approach. You will find the letter below.
In this article we focus on the wider issue of comparability in financial reporting. Many investors do not appear to appreciate how often financial statement data may not be comparable, the causes of that lack of comparability, and how this can impact investment decisions.
Comparability in financial reporting
Comparability (or the lack if it) in financial reporting is a significant issue for investors. Whatever the type of investor or the approach to making investment decisions, financial statement data invariably plays an important part of the process. If that data is not comparable, investment decisions may be sub-optimal, with negative consequences for capital allocation in the wider economy.
Comparability of financial statement data is important for all types of investors
The need for comparability is perhaps most obvious where investors use quantitative analysis, such as in factor investing or in initial screening before more detailed fundamental analysis. Such analysis typically focuses on headline metrics such as earnings, leverage and return on investment, with limited, if any, adjustments made to reported data. If that data is not comparable, the factors may be misrepresented1We examined the value factor and the impact of inconsistent intangible asset accounting in our article ‘Intangible asset accounting and the value false negative’. and the screening of limited use.
Where more detailed analysis is done, and investors delve into the financial statements in more detail, there is more opportunity to allow for accounting differences, including adjusting to produce comparable metrics. However, such adjustments can be difficult or impossible, with the allowance for accounting differences being imprecise.
The need for comparability is recognised in the IFRS and US GAAP conceptual frameworks. In the case of IFRS, the framework includes this explanation:2From the IFRS conceptual framework paragraphs 2.24 and 2.25.
“ … information about a reporting entity is more useful if it can be compared with similar information about other entities and with similar information about the same entity for another period or another date. …. Comparability is the qualitative characteristic that enables users to identify and understand similarities in, and differences among, items.”
Comparability refers to both the ability to make comparisons over time for a single company and the ability to compare one company with another. A further aspect of comparability is that companies should account for economically similar transactions in the same way (which is part of the problem with credit losses). If any of these aspects are compromised, so too is the understandability and usefulness of financial statements.
The challenge for investors is that while a lot of accounting data is highly comparable, this is not always the case. A lack of comparability can arise from different sources, including:
- A difference in accounting standards, such as IFRS versus US GAAP
- Options within accounting standards
- The application of judgement
- Differences in the interpretation of ‘probability’ or ‘threshold’ words
- A lack of clarity of accounting standards or lack of guidance for certain transactions
- A lack of retrospective adjustments when accounting changes
We think it is important for investors to be aware of each of these sources of a potential lack of comparability. What follows is by no means comprehensive, but we hope it will at least help alert investors to the dangers of automatically assuming data used in making investment decisions is comparable.
IFRS versus US GAAP
In many respects IFRS and US GAAP are very similar, with several IFRS standards themselves based upon earlier US GAAP pronouncements. More recently the IASB and FASB actively worked together to reduces differences or, when working on new standards such as revenue recognition, to issue the same requirements. Unfortunately, the boards are no longer engaged in active convergence, and we have seen some more recent examples of new differences emerging where previously the requirements were the same.
At a detailed level there are many differences between IFRS and US GAAP. A useful summary of these, which we often recommend to investors, is provided by Ernst & Young: US GAAP versus IFRS – The Basics. In specific cases any one of these could be material, although most of the time they do not ‘move the dial’ for investors for the vast majority of companies. However, there are several areas where the differences have more widespread effects. Here are some examples:
Intangible assets
While the IFRS and US GAAP accounting for purchased intangibles, including those purchased as part of a business combination, is essentially the same, intangibles that are internally generated are treated quite differently. Under US GAAP any expenditure on research and development must be immediately expensed, whereas, under IFRS, expenditure on ‘development’ must be capitalised. IFRS capitalisation is subject to several criteria, but for many companies it results in the recognition of a significant asset, that would not appear in balance sheets under US GAAP. This capitalisation versus expensing also has an impact on the timing of the recognition of the related expense.
A good example is provided by the Autos industry. IFRS reporter Stellantis capitalises development expenses, as explained in their accounting policy note below. This has a material impact on both balance sheet, income statement and performance metrics.
Stellantis capitalisation of development costs
Stellantis financial statements 2022
The balance sheet amount of previously capitalised expenditure less accumulated amortisation is €15.7bn, which represents about 22% of shareholders’ equity. The amount capitalised in 2022 was €3.1bn whereas the amortisation of previously capitalised amounts in the same year is €1.6bn, a net benefit of €1.7bn, or about 9% of reported pre-tax profit, compared with a policy of immediately expensing development costs that is applied by US GAAP reporters.
For more about intangible asset accounting, and how the inconsistent and limited recognition of intangibles affects return on capital, see our article ‘Missing intangibles distorts return on capital’.
Lease accounting
From 2019 both IFRS and US GAAP were changed to require the recognition of (almost) all leases as a liability and a corresponding right-of-use asset in the balance sheet. While there are some differences in liability3One such difference concerns leases where payments are indexed to inflation – see our article ‘Beware the IFRS 16 inflation headwind’. and particularly right-of-use asset measurement, the balance sheet is reasonably comparable. However, the same cannot be said for the income statement, where for many leases US GAAP does not separate the interest and depreciation components of the lease expense. This has a significant impact on key performance metrics such as EBITDA and EBIT.
We discussed these differences in our article ’Operating leases: You may still need to adjust’ and show the numerical impact in ‘Interactive model: Convert US lease accounting to IFRS’.
Pension liabilities
Like lease accounting, the balance sheet impact of defined benefit pension liabilities of IFRS and US GAAP reporters are reasonably consistent, with the comparability issue once again affecting performance metrics.
One difference is the treatment of actuarial gains and losses (the effects of remeasuring pension assets and liabilities to current values at each balance sheet date). In both systems these are generally reported outside profit and loss in other comprehensive income. However, only in the case of US GAAP is the effect ‘recycled’ to profit and loss in subsequent periods. US GAAP also provides an option to report these items in profit and loss in the period the remeasurement occurs, which is not available in IFRS. (Although, because this item is generally omitted from non-GAAP metrics, the adverse impact on comparability is arguably reduced.)
A more important and often overlooked effect is how the pension financial income and expense is calculated. IFRS calculates an interest amount based on the net surplus or deficit whereas US GAAP incorporates a separate expected return on pension assets. The latter produces a more favourable profit and loss effect, with the impact being material for many companies with significant pension schemes.
We examined the impact of IFRS and US GAAP on apparent pension leverage in our article ‘Pension leverage under IFRS and US GAAP’.
Operating cash flow
Some differences between IFRS and US GAAP are presentational, but these are no less annoying for investors. A good example is in the cash flow statement.
Both IFRS and US GAAP present cash flows in operating, investing and financing categories, and reconcile to a bottom line of the change in cash and cash equivalents; but how these categories are defined when using the indirect format is different. Under US GAAP the operating category includes the cash flow effects of all gains and losses reported in net income, whereas under IFRS it is common for operating cash flow to be the cash flow equivalent of operating profit, with non-operating items reported as investing or financing cash flows. IFRS companies do presently have the option to use the US approach, but this will come to an end when the soon to be released IFRS Primary Financial Statements standard becomes effective – another example of IFRS and US GAAP further diverging (although we support the changes being made by the IASB).
Options in accounting standards
Some accounting standards give companies the option to select from different accounting methods. This could be a difference in measurement, when gains and losses are recognised, or in presentation. We have the impression that IFRS has more options than US GAAP. Some options in IFRS exist because when international standards were first developed there was an attempt to accommodate the different approaches of some jurisdictions. Subsequently accounting options were often provided to try and manage the very different views (often around the use of cost versus fair value measurement) of different constituents.
Hedge accounting is optional under both IFRS and US GAAP
One option that appears in both IFRS and US GAAP (and which may surprise investors) relates to hedge accounting. When companies use derivatives to hedge risks (including those related to currency fluctuations, commodity price changes and interest rates) recognition and measurement mismatches can arise. For example, the use of currency swaps to hedge the currency exposure in forecast revenues results in the fair value changes of the swaps being reported in profit and loss (if hedge accounting is not applied) while no gains and losses are reported in respect of the yet to be recognised revenues.
Hedge accounting is the process of adjusting the accounting to eliminate these mismatches – in this case by applying cash flow hedging and deferring the gains and losses for the currency swaps in other comprehensive income. However, the application of hedge accounting is optional. The accounting policy note below shows that IFRS reporter Rolls Royce applies hedge accounting to some transactions but not (“in general”) to the currency exposure for forecast purchases and sales – the example we describe above.
Rolls Royce hedge accounting disclosure
Rolls Royce 2022 financial statements
The application of hedge accounting by some companies, and not others, and its application to some hedges and not others by individual companies, is confusing and impairs comparability.
In the case of Rolls Royce, this lack of comparability is mitigated through their non-GAAP metrics where they adjust for the effect of their hedges. However, these adjustments are company specific and themselves may not be comparable with similar non-GAAP metrics provided by others that do not apply hedge accounting, or those companies that elect to apply hedge accounting as specified in IFRS 9.4There is a further option in IFRS accounting where companies can elect to apply the hedge accounting requirements specified either in IFRS 9, the most recent standard on accounting for financial instruments, or the previous requirements in IAS 39. The difference is unlikely to be material in most cases but nevertheless allowing two approaches seems unnecessary and inevitably confusing for investors.
The application of judgement
Investors often seem surprised by the amount of judgement that is required to produce financial statements. While much of accounting is a process of recording past transactions, many aspects of this require judgement, including how gains and losses are classified and allocated to the appropriate accounting periods. Furthermore, the measurement of assets and liabilities often involves more than a simple calculation of the transaction amount, with remeasurement at current values necessary to provide relevant metrics.
Many disclosures in financial statements are designed to inform investors about how judgements are made and of the key measurement inputs used. It is always worth noting the assumptions used to measure significant items in financial statements.
Interpretation of probability and threshold words
Some accounting standards differentiate between different accounting methods based on threshold terms that refer to either the probability of an event occurring or significance of an item. The number of such terms is surprisingly large and includes: substantially all, probable, highly probable, significant, insignificant, more than insignificant, and reasonably certain.
The problem is that most of these terms do not have a clear definition for the purpose of financial reporting. The exception is the use of ‘probable’ in IFRS, which means more likely than not (i.e. a greater than 50% probability). Unfortunately, the same term has a different, and less precise, meaning in US GAAP where it represents a higher level of probability, which is confusingly closer to the use of ‘highly probable’ in IFRS.
Loan impairment standards have thresholds that may be interpreted differently
Loan impairments provide a good example of the impact of threshold terms. Under IFRS the switch from a 12-month loan loss allowance to a full lifetime allowance takes place when there has been a significant increase in credit risk compared with the credit risk on the date the loan was originated. Because the word ‘significant’ is not defined by IFRS, it is left to companies to identify exactly how they implement the requirement based on their own internal systems and how they measure credit risk. Inevitably the way this ‘significant’ test is applied will differ and lead to a lack of comparability.
The US GAAP approach to loan impairments does not feature the same ‘significant’ test given that a lifetime allowance applies in all cases. However, a different threshold test applies to determine whether the gross-up method or the normal CECL lifetime loss approach is used for purchased loans. The gross-up method applies if there has been a ‘more than insignificant’ increase in credit risk since origination. The proposed change to US GAAP we explained in our last article would remove this threshold, although it introduces a different (and, in our view, still artificial) divide in distinguishing between originated and purchased loans and a new threshold of “seasoned”5‘Seasoned’ refers to the time between origination of a loan and its subsequent purchase by another company..
Here is how IFRS reporter Ping An describe the ‘significant increase in credit risk’ threshold, followed by their explanation about the judgements that are necessary to establish their provision for credit losses:
Ping An – Credit risk and critical accounting estimates footnotes extracts
Ping An financial statements 2022
Reading the above, the extent of management discretion over how threshold terms are interpreted may appear to be so great that it is difficult to see how any financial data can be comparable. However, in practice, the application of these thresholds may not produce as many comparability problems for investors as you might expect. Practice develops in financial reporting and the global coordination of auditing and securities regulation serves to produce generally accepted approaches to the implementation of seemingly vague requirements of accounting standards. However, if it appears that the approaches adopted by companies in dealing with thresholds seem to differ, then it may be worth you asking questions.
A lack of clear guidance or different interpretations of accounting standards
Although accounting standards cover most transactions, there are situations for which there is no specific guidance. Companies must then resort to developing an accounting policy based on their interpretation of the conceptual framework or by analogising to other accounting standards. Even where standards do exist, sometimes they may be unclear or incomplete. In these situations, there is further potential for a lack of comparability. Practice tends to gravitate towards a common approach and additionally the FASB and IASB have interpretation committees that help to reduce any diversity in practice. Nevertheless, differences remain, and investors may well be impacted.
We highlighted a good example of this in our article ‘When investors need to restate liabilities’, in which we discussed the discount rate used to measure long-term non-financial liabilities under IFRS, such as environmental or decommissioning provisions. In IAS 37 there is a reference to allowing for risk in determining the discount rate. There is, however, no clear explanation of what risk this is and whether it should include credit risk applicable to the liability. We understand that practice varies in this case.
The lack of clarity in IAS 37 is presently being considered by the IASB and hopefully should be fixed soon.
A lack of retrospective adjustment when accounting changes
When a company changes the accounting methods it uses, such as when a new accounting standard is first applied, the revised approach is generally applied retrospectively with assets and liabilities restated as though the new accounting had always been applied. Comparability is further enhanced by the restatement of the prior period results so that these are also based on the same accounting methods.
Unfortunately, this ‘fully retrospective’ approach is not always applied, often due to concerns about the cost or difficulty for companies to prepare the information. Any alternative approach (which should be clearly explained in the footnotes) has potential to impair comparability.
Historical restatement never extends beyond the one comparative period. Historical trends over several periods are often used in equity analysis and inevitably a discontinuity will arise when accounting practices change. Investors need to be careful when using growth rates or in making historical comparisons of metrics such as profit margins or leverage.
The following extract from the 5-year historical summary published by UK retailer Tesco includes two discontinuities. In 2021 the company changed their accounting policy regarding property buybacks – there is no indication of whether this significantly affects comparability. The second discontinuity occurred in 2019, from when the company applied IFRS 16 to capitalise most operating leases that were previously off balance sheet.
Tesco 5-year historical summary extract
Tesco financial statements 2022
The change in lease accounting is the primary reason for a significant increase in net debt from 2018 to 2019. Most companies would have simply noted that the change has taken place, to warn investors; Tesco helpfully went further. The company provides a separate disclosure of “Discounted operating lease commitments” in 2018. This is not exactly what the additional capitalised leases would have been had they applied IFRS 16 to that period, but it should be a reasonable proxy. This amount is included in their measure of “Total indebtedness” which does appear to be more comparable over time.
Accounting for credit losses – tell us what you think
Returning to loan loss reserves (the topic that prompted us to consider the wider issue of comparability in financial reporting), we are interested to hear your view.
We think the different accounting for credit losses for purchased and originated loans has no conceptual rationale, and that there is no valid reason why IFRS and US GAAP should differ at all in this area. Both differences impair comparability and adversely affect investors.
Submit your view to see the results of our poll so far.
Insights for investors
- Do not assume that financial statement data is always comparable, even if companies report under the same accounting standards.
- There are multiple differences between IFRS and US GAAP. Most will not materially affect investment decisions, but some have a more significant and widespread impact. Look out for differences in the accounting for intangible assets, leases and pension liabilities.
- Companies are sometimes able to choose between alternative accounting approaches, including differences in presentation and measurement.
- Financial statements depend on estimates and judgements. Where the effect could be significant, make sure you check the footnote disclosures to identify any potential comparability issues.
- Be careful when analysing bank financial statements – allowances for credit losses are not comparable due to the difference between the accounting for purchased and originated loans, and because of the differences between IFRS and US GAAP.