Accounting: not as black-and-white or as boring as you may think

Accounting: not as black-and-white or as boring as you may think

Mikhail Motala

Kevin Cousins

Mikhail joined PSG Asset Management in 2015 as an Equity Analyst. He conducts research on both local and global companies across various sectors. Before joining PSG, he worked in the assurance division at Ernst & Young. Mikhail is a qualified Chartered Accountant. Kevin is Head of Research at PSG Asset Management and has 23 years’ experience in investment management. After working at BoE Asset Management from 1993 to 2002, he co-founded Lauriston Capital, a specialist hedge fund manager. Kevin then worked as part of the hedge fund management team at Brait (now called Matrix Fund Managers). Kevin joined PSG as an Investment Analyst in 2015. Accounting is a nuanced business language that can reveal important clues The foundation of our investment process is research. Not just any research, but our own proprietary work, using companies’ audited financial statements as our main source of data.

These choices allow them to present a company in a specific way – usually a very flattering one. (Occasionally, however, before a ‘take-private’ transaction for example, incentives are reversed and the choices made will most likely be aimed at lowering reported profits.)

Investors seldom consider the accounting policies and disclosures behind a set of financial statements very intently. For most people, Generally Accepted Accounting Practice (GAAP) and its application is not an enthralling topic. However, this is the language of company reporting – and like any language, it is full of easily overlooked subtleties and innuendos. Investors who think accounting is boring may miss important clues that can help to avoid the very non-boring outcome of losing money.

While potentially masking true business performance (and making analysis more difficult), the choices themselves provide investors with valuable information. For example, a consistent pattern of aggression or conservatism in accounting policy and disclosure is vitally important evidence, providing a window into the true corporate culture of a business. In fact, we believe it provides better insight than the typically lengthy and buzzwordladen narrative at the start of most annual reports. It is equally important to scrutinise changes to policies and disclosures: what motivated the change and what is the impact on reporting?

Financial statements are finalised through negotiation Most investors are under the impression that the process of producing a set of accounts is precise, with independent auditors supervising management closely. This is not the case. Auditors do not verify that accounts are accurate, only that they do not contain ‘material misstatements’. In reality, the process of finalising accounts is one of negotiation (as detailed in Table 1) and important clients (who pay substantial fees) can have strong negotiating positions. Accounting choices can obscure true performance but still (ironically) provide valuable insight Management teams have far more discretion in their choice and application of accounting policies than generally imagined.

Case study 1: Comparing African Bank’s and Capitec’s bad debt provisioning policies When the accounting policies of companies in the same industry are vastly different, it often raises red flags. We took African Bank’s accounts from 2009 to 2013 (the last set of accounts it published before its failure in 2014) and recalculated its provision for bad debts in every year using Capitec Bank Holdings’ (Capitec's) provisioning policy. Of course, while African Bank and Capitec both focus on unsecured lending, no two banks are alike. Differences in their provisioning policies may therefore partially be due to the different compositions and durations of their lending books. But even given these limitations in this relatively simplistic exercise, the results make for interesting reading.

Table 1: Accounting perception versus reality Investors’ perceptions

How it works in reality

Management must follow strict accounting rules and conventions when presenting a company’s financial statements.

Management has plenty of discretion in how it presents its accounts.

The accuracy of financial statements is verified by an independent audit.

The final set of accounts is often the result of an intense negotiation between auditors and management over ‘audit differences’.

Large blue chip companies produce better accounts, with less risk of misleading policies or inadequate disclosure.

Larger firms often pay significant fees to auditors, and therefore have much more negotiating power. The professional management teams typical of large firms often have incentives that result in the unintended consequence of promoting misleading accounting or disclosure.

Source: PSG Asset Management

10 |

Graphs 1 and 2: African Bank – net book value and earnings per share (2009 - 2013) Net book value (NBV) per share (rands)

Earnings per share (EPS) (rands)

20

4

15

3

10

2

5

1

0

0

-5

-1

-10

-2 2009

2010

2011

2012

2013

-R5.29 2009

2010

2011

2012

2013

African Bank reported NBV

African Bank reported basic EPS (2013 truncated)

African Bank PSG-adjusted NBV

African Bank PSG-restated EPS (2010 onwards)

Sources: African Bank annual financial statements, PSG Asset Management research

In Graph 1, the gold bars show African Bank’s reported net book value (NBV) per share, while the charcoal bars are our estimates of NBV if African Bank had consistently adopted the same bad debt provisioning policy as Capitec. While our estimated NBVs are much lower, the trend itself also provides interesting information – our figures decline over time while African Bank’s reported numbers initially increased. In the year to September 2013, African Bank took a substantial write-down on its lending book (hence the decline in its NBV). However, by this date our adjustments resulted in a negative value, indicating that African Bank would have been insolvent. As we know, despite raising some R5.4 billion of additional capital in a rights offer in December 2013, African Bank did go under less than a year later. The earnings per share (EPS) shown in Graph 2 reveal a similar picture. Our re-calculated numbers (in charcoal) are a fraction of African Bank’s reported EPS and do not show the growth trend illustrated between 2009 and 2012. (The validity of this trend as a fair representation of earnings is called into question anyway by African Bank’s own write-offs in 2013 and its subsequent collapse in 2014.) Importantly, at the end of 2012, African Bank traded at an attractive seven times earnings on its reported EPS. By using our restated EPS, the price-earnings ratio increases to a heady 56 times.

Of course, the analysis is easy in hindsight. However, this exercise reveals that important clues about African Bank’s true health were available from at least 2011, and reinforced when its 2012 results were published in November 2012. As Graph 3 shows, despite the share price being substantially lower than previous highs, a sale in late 2012 would have preserved the majority of an investor’s capital. Management can be very good at getting investors to ignore statutory earnings Many management teams disclose their own adjusted earnings alongside statutory numbers, using disclaimers such as ‘adjusted’, ‘normalised’, ‘underlying’, ‘operating’ or ‘nonGAAP’. The idea is to eliminate once-off or non-cash items that distort reported earnings, thereby providing ‘clearer’ disclosure. While improved disclosure is laudable, there are several potential problems with adjusted earnings. Firstly, the adjustments are subjectively made by management, and are not subject to accounting standards or audit opinion. As time goes by, the temptation to classify more expenses as ‘non-operating’ grows, especially in tough business environments when it is likely that a company will miss management’s earnings guidance. This temptation is further compounded if management incentivisation is based on adjusted earnings rather than statutory earnings.

THIRD QUARTER 2017 | 11

Graph 3: African Bank’s share price (rands) (2009 - 2014) 30 25 20 15 10 5 0 AUG '09

FEB '10

AUG '10

FEB '11

AUG '11

FEB '12

AUG '12

FEB '13

AUG '13

FEB '14

AUG '14

Source: Bloomberg

Case study 2: IBM – statutory versus operating earnings International Business Machines Corporation (IBM) had a long track record of beating guided earnings, both quarterly and compared to its long-term roadmaps, which targeted growth over specific five-year periods. The first roadmap ended in 2010, with IBM consistently beating forecast EPS. When the company announced its 2010 results, IBM’s management also announced a new target for the next five years – achieving $20 of EPS in 2015. However, there was one change to the previous roadmap: the $20 would be ‘operating’ EPS, rather than statutory reported EPS.

There were several other factors that could have indicated the growing risks: • Free cash flow dropped substantially below reported earnings from 2013 onwards, compared to a history of producing earnings largely in cash. • The CEO and CFO both retired, with new internal appointments in these roles in October 2012 and January 2014 respectively. • The new CEO’s incentive remuneration for hitting targeted earnings dwarfed what she could have made from dividends and price appreciation on her IBM stock holdings.

By mid-2014 IBM had beaten its guided earnings estimates for 31 quarters in a row. This was an astonishing performance considering how the vagaries of different economies, currencies and the timing of customer orders – which are all outside management’s control − can impact any single trading quarter. In October 2014, IBM shocked the market by missing earnings. In addition, it announced it would not make its roadmap target of $20 of operating EPS in 2015. The share price dropped by 35% over the next 18 months.

Management’s accounting policy choices can provide valuable insights and should not be overlooked Investors should understand that management teams often have considerable leeway in their choice of accounting policies and disclosures. These choices also give good insight into a company’s corporate culture. It is important to carefully evaluate non-statutory earnings numbers and incentives that drive management behaviour, as they are critical in determining accounting risk.

Were there any accounting clues that could have warned investors that IBM’s run of exceeding management’s forecasts was coming to an end? Graph 4 shows that, within three years of changing its roadmap target from statutory EPS to operating EPS, the adjusting items had grown from a negligible amount to just under $1.5 billion.

An in-depth knowledge of accounting and the process of producing financial statements is a huge asset in identifying the often subtle signals that trouble may be brewing in a company. We therefore believe it is essential to have analysts in our team who, in addition to having excellent technical ability, do not regard accounting as boring.

12 |

Graph 4: IBM operating earnings adjustments after tax ($m) (2009 - 2015) 1 600 1 400 Incentive system changes to exclude items

1 200 1 000 800 600 400 200 0 2009

2010

2011

2012

2013

2014

2015

Sources: IBM 10-K annual reports, PSG Asset Management research

THIRD QUARTER 2017 | 13