This is a transcript of a key-note address that Steve gave to a conference organised by the UK Endorsement Board, the body that endorses IFRS for use in the UK and provides UK input into the development of IFRSs.
Steve’s 30 year journey from analyst to standard setter, and now as author of The Footnotes Analyst, covers a period of dramatic changes in financial reporting. The degree of comparability and the quality of financial data available to investors today is infinitely better than he faced back in 1997.
Note: The original text of the speech has been lightly edited for clarity and to make it flow better in a written format. The views expressed are Steve’s own and should not be associated with any organisation for which he has previously worked.
Thank you, Seema, for that introduction. It is great to be here.
I have followed what the UK Endorsement Board has been doing for several years and have read much of your research. I have even dialled-in to listen to some of your Board meetings. That may sound rather sad for a retired standard-setter, but once you become interested in standard setting and financial reporting, it is difficult to let go. You become invested in improving the quality of reporting, and that has certainly happened to me. I have caught the standard-setting bug.
Congratulations on your first five years, and congratulations on this conference. I am particularly looking forward to the discussions later on intangibles and cash flows.
When I spoke to Seema and Paul about this session, they suggested that I reflect on my time both as a standard-setter and as a user of financial statements. That is exactly what I intend to do. I am going to take you back to the period before the IASB and use that perspective to highlight the importance of global, high-quality accounting standards—which, in a sense, is why we are all here today.
The UK has been highly influential in that journey. The first meeting of the IASC took place here. The work of the former Accounting Standards Board and now the UK Endorsement Board, has been influential. The IASB itself is based in London. The UK has always punched above its weight when it comes to accounting and standard setting. We may not always be able to say that about other areas, but I think we can certainly still say it about financial reporting.
What I want to discuss today is what makes financial statements useful for investors. In my view, there are two key ingredients:
- Comparability – the ability to compare financial information across companies and over time.
- Quality – faithful representation, relevance, understandability, timeliness and the other characteristics set out in the Conceptual Framework.
To explore both themes, I shall take you back nearly 30 years to 1997, when I first joined UBS and became deeply involved in international accounting. I will then bring the story forward to the present day. My purpose is partly to celebrate the extraordinary progress that has been made – looking back to 1997, the financial reporting landscape was dramatically different from what we have today. However, there is still more to do, and I will also make a few suggestions for future developments.
Comparability
Comparability is fundamental for investors. Analysts are constantly comparing one company with another and investors must choose between competing investment opportunities and construct portfolios from a vast range of options. Even when analysing a single company, comparable information remains essential because it allows performance to be assessed over time and against peers.
Nearly 30 years ago I joined UBS as an equity analyst, having previously delivered training programmes for the bank’s analysts and corporate finance professionals, and contributed to a research project on equity valuation. My new role had two dimensions: helping UBS analysts with their work and engaging externally with clients on accounting and reporting issues. However, there was one particular challenge they wanted me to tackle – developing comparable data to support a global research product.
At the time, investment research remained largely organised by country. There was a UK analyst team following UK companies, a German team in Frankfurt covering German companies, and so on. But the industry was increasingly moving towards regional and global sector teams.
UBS wanted global data such as earnings, price-to-earnings ratios, EV/EBITDA multiples and all the other metrics investors rely upon. The problem was that the available data was fragmented and, importantly, not comparable
They asked me to help build a global, comparable database. Naturally, I wanted the job, so I said: “Of course. No problem.” But clearly, in reality, it was impossible. I did make some progress: I standardised certain terminology. In the UK template, “stocks” became “inventories”. I developed a definition of operating profit—which, interestingly enough, the IASB has now formalised almost 30 years later.
I presented a global template for research data and everyone seemed pleased with it. What I did not emphasise was that beneath the template, recognition and measurement were completely different from one jurisdiction to another. The appearance of comparability was there; the reality was not.
Since 1997, however, we have come an extraordinarily long way. When I joined UBS, IFRS did not yet exist in its current form. Today, we have a series of standards that have transformed comparability across capital markets.
IFRS 18, for example, represents a major improvement in financial statement presentation. Simply having a defined operating profit measure is a significant step forward for investors.
I remember telling investors back in 2003 that help was on the way regarding presentation comparability. At the time, I invited IASB staff members working on what was called the Performance Reporting project to present to UBS clients. They had developed something known as the Matrix Approach, which I thought had enormous potential. I told clients not to worry: in a couple of years they would finally get a meaningful and comparable operating profit measure. That prediction proved somewhat optimistic. Nevertheless, after several iterations of the project, we have eventually arrived at IFRS 18, a standard that has been particularly welcomed in Footnotes Analyst articles.
Similarly, IFRS 17 has delivered, for the first time, genuinely comparable insurance accounting across jurisdictions. It was a long journey, and I am proud to have played a part in it.
IFRS 15 has had a comparable effect on revenue recognition. Before its introduction, analysts often faced major challenges in understanding how companies recognised revenue. Disclosure alone was rarely enough to make meaningful adjustments for differing accounting policies. I remember analysing a franchising arrangement used by a food retail company that appeared fundamentally different from industry practice. Despite considerable effort, it was almost impossible to determine precisely what was going on or to make reliable adjustments. IFRS 15 addressed many of those problems and substantially improved comparability.
And I should not overlook IFRS 16. I will return to leasing later, but it too represents an important milestone in the development of global accounting standards.
Comparability: Progress and Remaining Challenges
We have come a very long way in terms of comparability, but there remain areas where further progress is possible.
One of the great disappointments during my time at the IASB was the failure to achieve convergence with US GAAP. For a period, it promised a great deal. A few months after I joined the Board, I remember listening to the SEC meeting at which it approved the removal of the requirement for foreign private issuers to reconcile IFRS financial statements to US GAAP. That was a landmark moment in the development of both IFRS and the IASB.
We also had the Norwalk Agreement. We held joint Board meetings virtually every month. For a time, we had joint project teams working together. There was a genuine sense that we were moving towards a single global accounting language.
Ultimately, however, that ambition was not realised. Today, the relationship between the IASB and FASB appears to me to be more about monitoring each other’s activities than working towards common standards. Perhaps the current political environment is not conducive to convergence, but I still hope that one day we can move closer to a truly global reporting language.
IFRS has achieved a great deal. It is now used in well over 140 jurisdictions. However, it still represents only around 40 per cent of global market capitalisation. Most of the remainder (by market capitalisation) is reported under US GAAP. Significant differences therefore remain. I admit that I am not entirely impartial, but I believe IFRS, and not US GAAP, has reached the better answer in several important areas.
Leasing
Take leasing as an example. I believe the IASB got it right.
The FASB approach continues to recognise a debt-like liability on the balance sheet while reporting a single lease expense in the income statement for many leases. To me, that lacks conceptual consistency. If there is a financial liability on the balance sheet, it should give rise to interest expense. If there is an asset, one would expect depreciation.
The IFRS approach provides a more faithful representation of the economics.
Loan Impairments
Loan impairment accounting is another area where significant differences remain.
Looking back, I think the FASB’s Current Expected Credit Loss (CECL) model took the wrong path. The front-loading of expected losses through immediate recognition strikes me as less persuasive than the IFRS approach.
One lesson I took from the development of IFRS 9 is that standard-setting bodies must remain focused on the needs of investors and on producing the highest-quality financial reporting. During those discussions, banking regulators exerted considerable influence. I remember attending meetings where the conversation often appeared to have a prudential flavour rather than a financial reporting one.
That distinction matters. The objectives of prudential regulation and financial reporting are not necessarily the same, and standard-setters need to maintain a clear focus on providing useful information to investors.
To be fair, the FASB has recently moved somewhat closer to the IFRS position in certain areas. Some of the developments in gross-up method have narrowed differences at the margin. Nevertheless, significant divergence remains.
Convertible Bonds
Another area where accounting differences have widened is convertible bond accounting. This may sound like a niche topic, but it is becoming increasingly important. Issuance of convertible bonds continues to grow, particularly among technology companies financing significant investments in artificial intelligence and related infrastructure.
Recent changes in US GAAP have made bifurcation of convertible bonds much less common. As a result, a company issuing a zero-coupon convertible bond may report little or no periodic financing expense.
Under IFRS, by contrast, bifurcation typically results in a more meaningful recognition of financing costs. In my view, that outcome is considerably more sensible. This is one of those areas where IFRS and US GAAP were once relatively close but have now drifted further apart.
Unfortunately, once positions become established, they tend to become entrenched. I do not expect the leasing standards, for example, to be reconciled any time soon. Nonetheless, I hope standard-setters continue to recognise the benefits of global comparability and look for opportunities to reduce unnecessary differences whenever possible.
Accounting Policy Choices
A second challenge to comparability comes from accounting policy options.
One of the things I particularly like about IFRS 2 is that there are essentially no accounting policy choices embedded within it. The same is largely true of IFRS 15 and, in many respects, IFRS 18. However, other standards offer considerable flexibility. I would like to see a reduction in that flexibility where possible.
Take investment properties as an example. Entities may choose either the cost model or the fair value model. Yet if they use the cost model, they are still required to disclose fair value information in the notes. If fair value information is important enough to require disclosure, one might reasonably ask why it should not simply be recognised directly in the financial statements. From an analyst’s perspective, that would improve comparability considerably.
Other choices are more difficult to eliminate. One example is the fair value through OCI election for equity investments in IFRS 9. Originally, during the development of IFRS 9 this was intended to reflect business models and strategic holdings. However, it became an option available on an instrument-by-instrument basis, a completely free choice that reduces comparability across companies.
IFRS 17: Success and Compromise
IFRS 17 occupies a special place for me. The standard was issued just a few months before I left the IASB. I spent ten years working on the project, and the Board itself had been working on insurance accounting long before I arrived.
I believe IFRS 17 is a remarkable achievement. For the first time, we have a genuinely global accounting standard for insurance contracts. Yet it also illustrates the challenges that arise when trying to reconcile deeply entrenched views.
The standard contains a surprisingly large number of accounting policy choices. Out of curiosity, I recently asked AI how many there were. It initially suggested five to seven. The more examples I raised, the higher the number became. Eventually I reached eleven. That is probably slightly unfair, as some of those choices are really assumptions or estimates rather than accounting policies. Nevertheless, there are still many policy options.
Examples include:
- Whether to accrete interest on the risk adjustment.
- Certain approaches to allocating the contractual service margin (CSM).
- Whether to use present-value weighting for coverage units.
- The presentation choices relating to OCI.
The OCI choice is particularly significant. Companies can elect to recognise the effect of changes in discount rates in OCI, making corresponding choices for their investment assets. Different elections can result in significantly different reported outcomes, even when the underlying economics are similar. This can be very confusing for investors.
Over the last few years I have given many presentations on IFRS 17, and investors frequently ask: “Why do these options exist at all?” The answer is simple: politics. When we began the project, views among insurers were deeply divided. Some argued strongly for a historical-cost approach with locked-in discount rates. Others advocated a fully current measurement model using updated discount rates and market values. The starting positions could hardly have been further apart.
Against that backdrop, simply reaching a single global standard was an extraordinary achievement. The policy options represented part of the compromise that made consensus possible. Even so, I would welcome efforts during future post-implementation reviews to reduce some of those options. Eliminating even a small number would improve comparability and reduce complexity for users.
That does not mean I expect the major OCI election to disappear anytime soon. It is likely to remain for many years. But wherever we can simplify accounting and reduce optionality, I believe investors benefit.
The Importance of Investor Input
An important part of the standard-setting process is obtaining investor input, although that is often more difficult than it sounds.
During my time at the IASB, I spent a considerable amount of time meeting investors and discussing proposed standards. Their views are vital, but it is also important to recognise that investors, like everyone else, can become accustomed to existing practices and may sometimes resist change.
I remember one example during the development of IFRS 15 on revenue recognition. Mobile phone contracts were one of the most frequently discussed transactions. Under the previous accounting model, a telecommunications company would typically recognise a loss upfront when a handset was provided to a customer and then recognise profits over time as subscription revenues were received.
The industry often argued that the handset was merely a marketing tool and that the real profit arose from the service contract. IFRS 15 took a different view. The customer is purchasing two distinct goods or services: a handset and a telecommunications service. The accounting should therefore reflect those separate performance obligations.
I recall attending meetings involving telecommunications companies and analysts. Some analysts argued that they preferred the old accounting because it was closer to cash flows. Their view was that cash outflows occurred at the start and cash inflows followed over time. However, closeness to cash is not necessarily the same as faithful representation.
Financial reporting exists to portray economic activity, not simply cash movements. The lesson for standard-setters is that while investor views are invaluable, they must also be evaluated critically. The objective should always be to produce the most useful information, not necessarily to preserve familiar accounting treatments.
Transition Choices and Comparability
Accounting policy choices are not the only source of comparability issues. Transition provisions can also have long-lasting effects.
At first glance, transition choices may seem less problematic because their effects eventually disappear. In reality, however, some transition decisions can influence reported numbers for many years.
IFRS 17 provides a good example. The various transition approaches available when the standard was first adopted continue to affect reported results today. In particular, the fair value approach used to establish the contractual service margin will influence financial statements for years to come.
High-quality disclosures help users understand those effects, but comparability remains more challenging than it would be under a single approach.
Comparability of Estimates
Another challenge relates to estimates.
Modern financial reporting inevitably relies on estimation. There is no realistic way to eliminate that. However, there are instances where standards provide insufficient guidance to support consistent estimation.
Again, IFRS 17 offers an example. Consider the illiquidity premium included within discount rates. I have followed various industry surveys, including those published by PwC, and it is striking how wide the range of assumptions remains across insurers. That is perhaps unsurprising because there is no universally accepted theoretical method for determining an illiquidity premium.
Some variation is inevitable. Nevertheless, where estimates continue to produce very wide differences in practice, it is legitimate to ask whether the underlying approach is sufficiently operational. At some point, persistent diversity may suggest that a different solution would provide more useful and comparable information.
Notably, the US insurance accounting model avoids the need for an illiquidity premium altogether.
Quality of Financial Reporting
Comparability is critically important, but quality matters just as much. By quality, I mean faithful representation of economic reality and to illustrate that point, I again return to 1997.
When I joined UBS, the FASB had recently issued FAS 123, dealing with share-based payments and employee stock options. At that stage, the standard did not require companies to recognise an expense. Instead, it required disclosure of what the expense would have been.
The reason was straightforward. Intense lobbying from technology companies had prevented mandatory expense recognition.
For analysts, however, the new disclosure was a revelation. For the first time, investors could see the economic significance of share option grants. What had previously been largely invisible suddenly became observable.
I spent much of the following years analysing these disclosures and writing research reports. The evidence was compelling. Share options represented a genuine cost to shareholders and, eventually, IFRS 2 established mandatory recognition of that cost.
I regard IFRS 2 as one of the IASB’s greatest achievements. It represented a considerable improvement in the quality of financial reporting and overcame significant opposition from preparers.
At one stage, under US GAAP, companies could choose whether or not to recognise an expense. Almost none did. Out of the S&P 500, only a handful elected to do so. Moving from that position to one where expense recognition became mandatory represented a major advance in faithful representation.
Defending Past Achievements
One lesson from the stock-option debate is that progress cannot be taken for granted. A few months ago, I read an article arguing that stock option expense recognition was unnecessary because dilution already captures the economic effect. The implication was that recognising both dilution and expense amounts to double counting. To me, that reflects a return to arguments that were thoroughly debated many years ago.
Standard-setters cannot focus solely on future projects. They must also defend and preserve past achievements. Otherwise, there is always a risk of moving backwards.
Pension Accounting: An Example of Long-Term Progress
Another area that has undergone remarkable improvement is pension accounting.
When I first studied accounting, there was effectively no meaningful pension accounting standard. Contributions paid into a pension scheme were recognised as an expense, and little else appeared in the financial statements.
Subsequent standards represented gradual improvements.
SSAP 24 introduced a framework for pension accounting, but it relied heavily on actuarial smoothing techniques and actuarial asset values that often bore little resemblance to economic reality.
The UK FRS 17 was a major step forward. It required pension assets and liabilities to be recognised more transparently and generated significant controversy at the time. Many viewed it as being hostile to defined-benefit pension schemes whereas, in reality, it simply revealed their economic consequences.
Later developments, including IAS 19, and the eventual elimination of the corridor method, further improved transparency. Today we have pension accounting that is vastly superior to what existed several decades ago.
Ironically, these improvements arrived just as many defined-benefit schemes were entering long-term decline. Nevertheless, the journey illustrates how accounting standards often improve incrementally over many years.
Looking Back—and Looking Forward
Standard setting is never purely technical: it is a political process as well as a technical one. Of course, I would prefer it to be as technical as possible and as focused as possible on the needs of investors. Yet standard-setting takes place in the real world, with competing interests, lobbying pressures and practical constraints.
Even so, we should recognise how far we have come. If someone had shown me today’s reporting landscape back in 1997, I would scarcely have believed it. Today’s analysts take for granted many things that simply did not exist thirty years ago:
- Lease liabilities on balance sheets.
- Comparable revenue recognition.
- Recognition of share-based payment expense.
- Global insurance accounting standards.
- Consistent operating profit measures.
These achievements represent the collective efforts of many people, including many people in this room.
That does not mean there is nothing left to improve. I would still like to see fewer accounting policy options. I would like greater convergence internationally. I would like improvements in areas such as convertible bond accounting and other outstanding issues. However, while continuing to improve financial reporting, we should never lose sight of how much has already been achieved. The progress of the last thirty years has been extraordinary.
It is also a reminder of why participation matters—whether through writing comment letters, serving on advisory groups, participating in research projects or engaging in conferences such as this one. Those activities are fundamental to achieving the shared objective of high-quality, globally comparable financial reporting standards.
Conclusion
Much has been achieved, but much remains to be done. I look forward to the discussions later today on cash flows and intangibles—two topics that have always been particularly close to my heart.
Thank you very much and I would be delighted to take any questions.