March 6, 2012

Sub Sahara|Kenya|Banks

Industry View
CONSTRUCTIVE

KENYAN BANKS
NPLs growth<Loan growth>GDP growth higher risks: A critique on credit risks
StanChart 175 262 -33.2% HOLD HOLD 164.69 -5.9% 8.1% 2.2%

Recommendations
Barclays Coop Current price Price at IoC Loss since IoC Current rec. Rec. at IoC FY12 TP Potential Dividend yield Total potential 13.85 16.7 -17.1% HOLD HOLD 14.57 5.2% 8.3% 13.4% 11.95 18.25 -34.5% BUY SELL 14.60 22.2% 6.0% 28.2% Equity 19.00 24.75 -23.2% BUY BUY 22.88 20.4% 6.6% KCB 21.25 22.25 -4.5% BUY BUY 24.89 17.1% 9.6%

27.0% 26.8%

Whats changed?
EPS Forecasts 2012F OLD NEW 2013F OLD NEW 2014F OLD NEW Target Prices OLD NEW Barclays 1.65 1.64 1.81 1.77 N/A 1.95 14.01 14.57 KCB 3.58 4.08 4.19 4.68 N/A 5.27 22.78 24.28

Stock Performances
YTD, local currency return
KCB 26.1%

Equity

15.9%

NSE ALSI

9.9%

StanChart

9.4%

Barclays

6.1%

Coop -2.4%

-5.0%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

System Loans vs. NPLs

Net loans 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Total NPLS

Peter Mushangwe +27 11 551 3675 peterm@legae.co.za

Mind the gap - System assets growth outperform economic growth: A primary macro issue in system credit risks analysis is the rate of system assets growth vs. its supposed natural long-term growth (nominal GDP growth). In most Sub Sahara African (SSA) systems, the low penetration levels imply stronger banking assets growth vs. nominal GDP. Nevertheless, banking assets growth should decline as penetration increases, and in Kenya the strong penetration into the retail segment in particular, points to possible narrowing of the gap that has expanded strongly in the past decade. Defying gravity - Loans outpace Non-Performing Loans (NPLs) growth by a colossal margin: While loan growth has materially outpaced nominal GDP growth, NPLs growth, contrary to logical expectations, has meaningfully lagged loan growth. Between 1H06 and 3Q11, system loans and advances have expanded by 2.9x, while system NPLs have declined from Kes102bn to Kes57.7bn. Rationally, one would expect the two, particularly in Kes-terms, to move together, not in lock-step, but generally in the same direction. While most investors are concerned by the impact of high interest rates to banks trading portfolios, a medium to long-term and more disturbing concern to us is this divergence in loans and NPLs growth rates. Are we cynical - Could the NPLs be gamed? Faster writing-off, restructuring of loans and conservative and less objective approaches to identification of NPLs are some of the ways the NPLs can be concealed. However, largely it depends on the regulators ability to ensure abidance by the prudential guidelines. In our view, the Central Bank of Kenyas (CBK) NPLs recognition and provision guidelines are fairly tight. That said, the choice to report under IAS39 means management discretion remains. The hungry and the satisfied - Local banks have been growing market share: Against the backdrop above, local banks, particularly Equity, Cooperative and KCB have been gaining significant market shares in the system. Equity which was #10 by assets with a market share of 1.9% in CY05 is now #5 with a market share of 8% (CY10), for example. Meanwhile Barclays, Stanchart and Citi have shed off assets market shares from 17% to 10.3%, 11.8% to 8.5% and 5% to 3.7% respectively between CY05 and CY10. Of course, gaining market share per se does not imply acquiring clients from competition as penetration and number of accounts in the system has increased, but it is an opportunity costs to losers.

Mar-

Mar-

Mar-

Mar-

Sep-06

Sep-07

Sep-08

Sep-09

Sep-10

Mar-

Dec-06

Dec-07

Dec-08

Dec-09

Dec-10

Sep-11

Jun-06

Jun-07

Jun-08

Jun-09

Jun-10

Jun-11

Who is swimming naked Cooperatives NPLs have reduced meaningfully despite a bloating loan book: Cooperatives loan book has expanded from Kes28.4bn in CY06 to Kes86.6bn in CY10, a 25% Compounded Annual Growth Rate (CAGR). Yet the credit costs have reduced from Kes1.4bn to Kes799mn (-38.4% CAGR) over the same period. This compares unfavourably to the systems 17.2% loan CAGR and a 9.5% credit cost CAGR. It also flatters to Equitys loan CAGR of 48.3% and credit costs CAGR of 46.2% over the same period. While we are not being cynical, we are sceptical that Cooperative can continue to grow its loan book with credit costs declining. As at 3Q11, the NPLs ratio was in line with our universes average of 4.8%. Recommendations - Is the fundamental trend we notice a head fake? While we are concerned by the static NPLs in Kesterms (hence improving ratios), we doubt this analysis will inhibit the current positive momentum in banks shares. We continue to recommend investors to BUY KCB (new Target Price (TP) Kes24.89), Equity Bank (TP Kes22.9) and Cooperative bank (TP Kes14.6) while we HOLD Barclays and StanChart. Our top pick is KCB. We believe KCB boasts a better balanced portfolio between retail and corporates, lower valuation risks and better opportunities to enhance asset quality vs. Cooperative and Equity. Improving efficiencies and regional profitability are key catalysts in the medium term (FY11-FY14), in our view. A note on KCB FY11 results Outperformed our aboveconsensus forecast: KCB has produced a strong FY11 set of results that beat our earnings per share (EPS) estimate of Kes3.14 and consensus of Kes3.01. Compared to our forecasts, the stronger deposit growth (32% vs. Legae estimate of 20%) led to higher loans and advances (Kes198.7bn vs. Legae estimate Kes189.1bn). Consequently interest income came higher at Kes27.9bn vs. our forecast of Kes27.6bn. Net interest income was 2% higher at Kes23.3bn vs. our expectation of Kes22.9bn (volume outweighed falling margins given our lower loans and advances forecast). Non-interest income also beat our forecast at Kes9.18bn vs. our estimate of Kes7.89bn while credit costs came in line at Kes1.89bn as per our expectation (an outperformance though considering our lower loan and advances estimate). Our new EPS for FY12 is Kes4.08, a 10% growth to FY11 as we expect the high base to restrain growth. Nonetheless, our EPS forecast is 12% above consensus. Management yet to grant us audience; Coop, Equity and StanChart yet to release: While management has promised to grant us an opportunity to discuss the results with it, we find frustrating that it has taken this long since results release date. Coop, Equity and StanChart are yet to issue their FY11 results hence comparisons are mainly based on 3Q11.

Page 2 of 22

Table of Contents
1. A look at the system: Always mind the gap Has system credit expanded excessively vs. GDP? Loan growth vs. NPLs growth: Defying force of gravity? Are we being cynical? Can NPLs be gamed? Are the high coverage ratios valuable? 2. A look at our universe: Who is swimming naked? Who has been taking more risks... ...and who is swimming naked? Forecasts, Recommendations and Price performances 3. KCB FY11: Outperformed our above-consensus EPS 4 4 6 8 10 14 14 16 17 20

Page 3 of 22

1. A look at the system: Always mind the gap

We have taken some time to look at the development of the Kenyan banking systems credit risks indicators. We admit that the picture is highly pleasing, with most possible metrics used in identifying risks showing improvements year after year. In this report, we put forward a few questions to investors. We attempt to answer them as well. Has credit expanded excessively relative to the economy? As Fig 1 below shows, the system assets have expanded by 3.3x vs. 2.6x for nominal GDP. In our view, this strong outperformance is chiefly a response to the low penetration levels. As a result, while it should generally indicate rising credit risks, (i.e. credit growing exceedingly faster than the capacity of the economy) investors could give the banks/system the benefit of the doubt. But again, the explosive growth in the retail segment, particularly by the big local banks, evokes worries. The number of accounts in the system has increased to 11.9mn (CY10) from a mere 1.7mn in CY02. The deposit accounts/bankable population ratio is now ~63%, despite the modest banking assets/GDP ratio of <60%. In our view, penetration particularly in the retail space has increased considerably in recent years; thus limiting future growth and increasing credit risks. (see Fig 2). This strong growth is also indicated by the high average credit multiplier of 1.8x i.e. system assets expand ~2x nominal GDP growth. (see Fig 3). As penetration increase, the multiplier will natural moderate. But even if penetration was to remain low, the gap between nominal GDP growth and credit growth will become a key macro-prudential metric in future. The countercyclical buffer proposed by the Basel Committee requires the monitoring of the gap between nominal GDP growth and credit growth in a system. A strong positive gap (GAPt = Credit/GDPt Trendt) requires additional capital as excessive credit risks are assumed to build in the system. The objective is to ensure that credit does not grow or contract excessively relative to GDP. We note that after a steep decline in system assets growth CY10, our proxy of the gap has turned negative despite enjoying a robust positive gap between CY05 and CY09. Credit growth has now strongly recovered. Kenyan regulators are unlikely to implement Basel 3, but we believe the gap is a novel tool that macro prudential regulators should keep an eye on. Would relatively high interest rates tame the growth? Intuitively, high interest rates should tame loan growth. Looking from CY07, interest rates were increasing, and so was the loan growth. (see Fig 4). We admit, the time period could be too short to bring out insights that may not be time-period specific, but prima facie, co-movement between interest rates and loan growth does not look highly negatively correlated. Poorer transmission mechanisms (of the monetary policy) are the chief reasons that lead to a significant lag.

Page 4 of 22

Fig 1: Notwithstanding the lower penetration argument, industry assets growth has significantly outperformed economic growth and its trend, particularly between CY05 and CY09...
Industry assets 4.0 Nominal GDP 0.60

Assets/GDP ratio and the Gap

3.5 3.3 3.0 2.6 2.5

0.55

0.50

2.0 0.45 1.5

1.0 0.40 0.5

0.0

0.35

Fig 2: ...meanwhile penetration has increased meaningfully, especially in the retail space

Fig 3: Asset growth declined faster than GDP in CY10. The average credit multiplier is ~2x. As the credit growth recovers, the multiplier should revert to its mean.
Industry 30.0% 8.0 25.0% 7.0 6.0 5.0 15.0% 4.0 10.0% 3.0 2.0 5.0% 1.0 0.0% 0.0 -1.0 GDP 9.0 Credit multiplier average

2000
20.0% -5.0% 2001

Number of deposit accounts Deposit Accounts/Population Deposit Accounts/Bankable population est. Number of system branches

2002

2001

2002
2003

2003
2004

Source: CBK, Legae Calculations. *Fin Access Survey est. bankable population of ~19mn

2004

2005

2005

Source: CBK, Legae Calculations

2006

Source: CBK, Legae Calculations

2006

2007

2007

2008
2008

2009
2009

2010
2010

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2002 1 682 916 5.2% n/a 446

2006 3 329 161 9.2% n/a 575

2008 6 428 509 17.2% n/a 887

2010 11 881 114 29.9% 62.5% 1 063

2001

2002

2003

2004

2005

2006

2007

2008

Page 5 of 22

2009

2010

Fig 4: High interest rates do not necessary tame loan growth in the short term. Poor transmission mechanisms to blame for the longer lag
Loan growth, LHS 45.0% 40.0% 35.0% 30.0% 14.0% 25.0% 13.5% 20.0% 13.0% 15.0% 10.0% 5.0% 0.0%
Dec-07 Dec-08 Dec-09 Dec-10 Sep-07 Sep-08 Sep-09 Sep-10 Sep-11 Jun-07 Mar-08 Jun-08 Mar-09 Jun-09 Mar-10 Jun-10 Mar-11 Jun-11

Lending rate 15.5% 15.0% 14.5% 50.0%

Change in loans

Change in av. lending rate

40.0%

30.0%

20.0%

10.0%

0.0% 12.5% -10.0% 12.0% -20.0%

Dec-07 Dec-08 Dec-09 Dec-10 Sep-07 Sep-08 Sep-09 Sep-10 Sep-11 Jun-07 Mar-08 Jun-08 Mar-09 Jun-09 Mar-10 Jun-10 Mar-11 Jun-11

11.5%

Source: CBK, Legae Calculations

Will NPLs continue to lag loan growth by such a colossal margin? Supervision reports highlight improvements in credit appraisal and monitoring: As we pointed out above, system credit growth has expanded in a vigorous manner in Kenya vs. economic growth. Logically, one would expect a worsening credit profile for the system, notwithstanding the low penetration. At best, NPLs growth should be close to loan growth. However, this is not the case in Kenya. In fact, NPLs have materially underperformed loan growth, and NPLs in Kes-terms are static at an average of ~Kes50bn, post a steep decline (~-40%) in CY07. Between 1H06 and 3Q11, the net loans of the system have grown by a meaningful 2.9x, while NPLs have declined by 0.57x, from Kes102bn in 1H06 to Kes57.7bn in 3Q11. Gross NPLs show a simple growth rate of -43% vs. a growth rate of 168% in Gross loans between 1H06 and 3Q11. As a result, the system NPLs ratio and to an extent the coverage ratios have substantially improved. The NPLs ratio has declined from 15.8% in 1H06 to 4% in 3Q11 a significant improvement by any measure in a period characterised by strong economic growth headwinds. (see Fig 5 Fig 6). Supervision reports have mainly attributed the decline in NPLs to improvements in credit appraisal and monitoring standards by the banks. Recoveries are also mentioned now and then.

Page 6 of 22

Fig 5: Loans have grown faster than the NPLs between 1H06 and 3Q11...
Net loans 3.5 Total NPLS 50.0% 40.0% 3.0 30.0% 2.5 20.0% 10.0% 2.0 0.0% 1.5 -10.0% -20.0% -30.0% 0.5 -40.0%
Mar- Mar- Mar- Mar- Mar-

Net loans

Total NPLs

1.0

0.0
Jun-06 Sep-06 Dec-06

-50.0%
Jun-07 Sep-07 Jun-08 Sep-08 Jun-09 Sep-09 Jun-10 Sep-10 Jun-11 Sep-11 Dec-07 Mar-08 Dec-08 Mar-09 Dec-09 Mar-10 Dec-10 Mar-11 Jun-11 Sep-11

Jun-07

Jun-08

Jun-09

Sep-07

Sep-08

Sep-09

Jun-10

Dec-07

Dec-08

Dec-09

Sep-10

Source: CBK, Legae Calculations

Fig 6: ...leading to improvements in most credit ratios.

Dec-10

Gross Loans Growth Net Loans Growth Gross NPLs Growth Gross NPLs/Gross loans Net NPLs Growth Net NPLs/Net Loans Total provisions Growth Specific provisions Specific provisions/Total provisions

3Q06 456.5 419.3 103.8 22.7% 66.6 15.9% 45.1 40.8 90.5%

3Q07 492.4 7.9% 476.4 13.6% 59.1 -43.1% 12.0% 43.1 -35.3% 9.0% 27.5 -39.0% 22.8 82.9%

3Q08 657.45 33.5% 644.2 35.2% 56.85 -3.8% 8.6% 43.6 1.2% 6.8% 27.3 -0.7% 22.6 82.8%

3Q09 736 11.9% 720.5 11.8% 60.6 6.6% 8.2% 45.1 3.4% 6.3% 24.8 -9.2% 19.4 78.2%

3Q10 878.8 19.4% 868.7 20.6% 61.2 1.0% 7.0% 51.1 13.3% 5.9% 34.3 38.3% 28.9 84.3%

3Q11 1192.5 35.7% 1181.7 36.0% 57.7 -5.7% 4.8% 46.9 -8.2% 4.0% 35.4 3.2% 27.7 78.2%

Source: CBK Legae Calculations

Page 7 of 22

Are we being cynical? Can the NPLs ratio be gamed? While there are various names for the credit costs that banks recognise in their profit and loss accounts, ranging from provision for bad debts to impairment charge, the key ratio investors often watch is the NPLs ratio. Unfortunately, this ratio can be gamed. While we are not being cynical and assume there is gaming of the NPLs in the Kenyan system we highlight that the NPLs can be concealed by: a) faster writing-off of NPLs (not necessarily a bad practice if the intention is not to game NPLs and it is a consistent policy). The steep decline in Kes-NPLs in CY07 seem to indicate a more radical write-off exercise in the system than normal (see Fig 7); b) restructuring of loans, particularly prior to classification as an NPL. This defers NPLs recognition, and is probably the most common practice by banks. NPLs can also be restructured although certain conditions should be maintained (ref Fig 9); c) slower disposal of security due to legal and other judicial issues yet the loan has been foreclosed. The NPLs would not be recognised if it is fully secured as it is moved to foreclosed assets for a period; and d) a conservative and less objective approach to recognition of NPLs. The IAS39 which requires impairments to be incurred if and only if there is objective evidence of impairment as a result of one or more loss events that have an impact on estimated cashflow on the financial asset or a group of assets that can be reliably estimated provides options to managers in their NPLs management. Of course the NPLs ratio can, and more often than not, decline due to improving asset quality. It could be argued that the declining Kes-terms value of suspended interest points to concrete credit quality improvements in the system as it is more difficult to game than NPLs. (see Fig 8). The caveat is that the lower interest rate vs. history could result in lower suspended interest irrespective of the credit risks.
Fig 7: Despite strong loan growth in CY07, system NPLs declined by >40% and so did the consequent provisions. In Kes-terms, system NPLs has remained largely static post the CY07 decline...
1.2 1.1 40.0% 1.0 0.9 0.8 0.7 0.6 -20.0% 0.5 0.4 -40.0% 0.3 0.2
Dec-06 Dec-07 Dec-08 Dec-09 Dec-10 Jun-06 Jun-07 Jun-08 Jun-09 Jun-10 Sep-06 Sep-07 Sep-08 Sep-09 Sep-10 Jun-11 Mar-07 Mar-08 Mar-09 Mar-10 Mar-11 Sep-11

Gross NPLs

Total Provisions

Specific provisions

60.0%

NPLs

Provisions

Specific provisions

20.0%

0.0%

-60.0%
Dec-07 Dec-08 Dec-09 Aug-07 Aug-08 Aug-09 Aug-10 Dec-10 Oct-07 Oct-08 Oct-09 Oct-10 Aug-11 Feb-08 Apr-08 Feb-09 Apr-09 Feb-10 Apr-10 Feb-11 Apr-11 Oct-11 Jun-07 Jun-08 Jun-09 Jun-10 Jun-11

Source: CBK, Legae Calculations

Page 8 of 22

Fig 8: ...hence with strong loan growth and static NPLs, ratios show material improvements. Suspended interest/Gross loans has also declined significantly although lower interest could have positively affected the ratio
Gross NPLs /Gross Loans 25.0% Total provisions/Net loans 9.0% 8.0% 7.0% 20.0% 6.0% 15.0% 5.0% 4.0% 10.0% 3.0% 2.0% 5.0% 1.0% 0.0%
Dec-06 Dec-07 Dec-08 Dec-09 Dec-10 Sep-06 Sep-07 Sep-08 Sep-09 Sep-10 Sep-11 Jun-06 Jun-07 Jun-08 Jun-09 Jun-10 Jun-11 Mar-07 Mar-08 Mar-09 Mar-10 Mar-11

Suspended interest/Gross loans

0.0%
Dec-06 Dec-07 Dec-08 Dec-09 Dec-10 Sep-06 Sep-07 Sep-08 Sep-09 Sep-10 Mar-07 Mar-08 Mar-09 Mar-10 Mar-11 Sep-11 Jun-06 Jun-07 Jun-08 Jun-09 Jun-10 Jun-11

Source: CBK, Legae Calculations

A note on renegotiated NPLs: Loan restructuring is one major contentious issue when interest rates increase as banks attempt to reduce their NPLs formation. Often, the duration of the loan is lengthened - a disadvantage to the bank if there is no corresponding increase in the interest rate charged - but a common way to ensure affordability by the borrower. When this restructuring is exercised prior a loan turning into an NPL, there is a higher chance the loan will remain classified Normal for longer than it should. However, the restructuring of NPLs is the one often monitored more stringently. The restructured/renegotiated NPLs are generally more important because reclassification would not only affect the provisions but the NPLs figure and the corresponding ratios when some of the NPLs are reclassified to Normal or Watch status. Fig 9 below shows the prerequisites for NPLs reclassification. In our opinion, CBKs required conditions for reclassification are stringent. We draw attention the fact that conditions (1) and (2) are explicit and precise, but condition (3) is subjective (and subject to abuse). This provides bank managers a leeway to reclassify NPLs as per their view. However, the CBK has authority to reclassify the loans, and when that occurs, an upgrade of the loan by a bank would require sufficient justfification. A note on write-offs: According to the prudential guidelines, write offs should be undertaken when the bank loses contractual control over the loan and the loan is deemed uncollectable and there is no realistic prospect of a recovery. The banks Board of Directors has ultimate authority for approval of the write-off. Writing-off also takes place only to facilities/loans classified as Loss and within 90 days, should there be no recoveries within the period. There are general indicators where writing-off a loan becomes evident e.g. when borrower becomes bankrupt; where all

Page 9 of 22

forms of securities or collateral have been realised but proceeds failed to cover the entire facility outstanding; where efforts to collect debt are abandoned for any other reason, among others. In our opinion, the last condition provides management with some discretion to determine, with the approval of the board, when to write-off NPLs. However, the guidelines are tight enough assuming the implementation is firm. We therefore do not expect abnormal write-offs in the system in pursuit of reducing NPLs.
Fig 9: The regulatory NPLs reclassification prerequisites tight enough in our view
NPL type
Substandard to Normal

Reclassified if and only if ...

(1) All past due principal and interest is repaid in full at the time of renegotiation (2) All past due interest is repaid in full at the time of renegotiation or (3) or a sustained record of performance under realistic repayment program has been maintained. All principal and interest payments are made according to the modified repayment schedule If condition (2) is sustained for 6 months, the loan can be reclassified as Normal; if condition (3) is maintained for 6 months, a Substandard renegotiated loan can be classified as Watch; The renegotiated loan will qualify as Normal if condition is sustained for 12 months (1) All past due principal and interest is repaid in full at the time of renegotiation

Substandard to Watch

and

Doubtful to Watch

Doubtful to Substandard or

(2) All past due interest is repaid in full at the time of renegotiation (3) or a sustained record of performance under realistic repayment program has been maintained. All principal and interest payments are made according to the modified repayment schedule If condition (1) is sustained for 6 months, the renegotiated loan can be reclassified as Normal; If condition (2) is sustained for at least 6 months, a Substandard renegotiated loan can be reclassified as Watch; and Normal if record sustained for 12 months; If condition (3) is sustained for 6 months, a Doubtful renegotiated loan can be reclassified as Substandard; reclassified as Watch if record sustained for at least 12 months; and reclassified as Normal if record sustained for 18 months.

and

Source: CBK, Legae Securities

The high coverage ratio may not be a sweet spot after all...: The systems coverage ratio is healthy at ~75%. There are two principal issues that we can deduce from the high coverage ratio; a) the system has more doubtful and lost loans than otherwise which therefore requires 100% provision as indicated by the prudential guidelines. The provisioning lag would also mean that the coverage ratio remains high even when new NPLs formation is declining due to the NPLs ageing and the consequent provisioning requirements. The NPLs of loans written between CY07 and CY09 are probably reaching Lost stage; b) NPLs are falling faster than provision. We have noticed that between 1H06 and 3Q11, NPLs in absolute terms declined by 49% (from Kes102bn to Kes57.7bn) while provisions have reduced by a lower amount of 21.5%. (from Kes45.1bn to Kes35.2bn). It can also mean both scenarios are playing out in the system. In addition, the low mortgage penetration (entails lower provisions as mortgages require less

Page 10 of 22

provisions due to better security/collateral values) means that the system carry higher provisions for loans, sustaining higher coverage ratio. Nonetheless, the provision/NPL ratio (coverage ratio) has remained fairly strong at an average of 65% between 1H06 and 3Q11. The specific provisions/total provisions ratio (i.e. specific provisions generally cover individually assessed/important impaired exposures) would imply an improving credit scenario in the system. (see Fig 10 - Fig 11). However, the general provisions/total provisions ratio has increased recently, which could signify that regulators are most likely having quiet words with bank managers and/or credit risk, particularly retail, has picked up materially. In all systems, regulators apply discreet pressure to bank managers in order to curtail risk propagation. Usually banks are asked to carry higher general provisions than they are ordinarily required to. We are unsure if this is the case in Kenya as moral suasion is highly secretive in banking and is extremely difficult to detect. However, as we have already indicated, the general provisions/NPLs ratio has increased to 16% in 3Q11 from 6% in 1H06. The general provisions/total provisions ratio has also increased to 22% from 9% over the same period. General provisions are made based on expected losses on undifferentiated pool of exposures such as credit cards and other smaller facilities. They should therefore, theoretically, grow as demand for credit increases, especially retail and SME credit prick up. They can also reflect managements loss expectations on the Normal loans. ...although we believe the regulatory provisioning guidelines are fair: As is the case in most SSA systems, a formulaic provisioning requirement is a key regulatory aspect in Kenya. Fig 12 shows the provisioning prudential guidelines. A 1% provision to Normal loans indicates the general provision that is forward looking to some extent. This is consistent with Basel 2 which requires a more forward looking provisioning practice. The IAS39 tend to be more conservative in recognising future NPLs (and therefore provisions), and according to the guidelines, the difference between the impairment charges computed under IAS39 and those required by applying the guidelines is appropriated off retained earnings and not expensed/credited to the profit and loss.

Page 11 of 22

Fig 10: The coverage ratio is high but specific provisions/total NPLs is only ~55%...
Provisions/Total NPLs
80.0% 70.0% 75.0% 65.0% 70.0% 65.0% 60.0% 55.0% 50.0% 50.0% 45.0% 40.0%
Sep-06 Sep-07 Sep-08 Sep-09 Sep-10 Dec-06 Mar-07 Dec-07 Mar-08 Dec-08 Mar-09 Dec-09 Mar-10 Dec-10 Mar-11 Sep-11 Jun-06 Jun-07 Jun-08 Jun-09 Jun-10 Jun-11

Specific Provisions/NPLS

Average

60.0%

55.0%

45.0%

40.0%
Jun-06 Jun-07 Jun-08 Jun-09 Jun-10 Sep-06 Sep-07 Sep-08 Sep-09 Sep-10 Jun-11 Dec-06 Mar-07 Dec-07 Mar-08 Dec-08 Mar-09 Dec-09 Mar-10 Dec-10 Mar-11 Sep-11

Source: CBK, Legae Calculations

Fig 11: ...as specific provisions/total provisions reduce (i.e. general provisions/total increase) along with loan growth
Loan growth 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% -10.0% -20.0% -30.0% -40.0% -50.0%
Feb-08 Feb-09 Feb-10 Feb-11 Apr-08 Apr-09 Apr-10 Apr-11 Jun-07 Jun-08 Jun-09 Jun-10 Aug-07 Oct-07 Dec-07 Aug-08 Oct-08 Dec-08 Aug-09 Oct-09 Dec-09 Aug-10 Oct-10 Dec-10 Jun-11 Aug-11 Oct-11

Total Provisions 95%

Specific provisions/Total provisions

Average

90%

85%

80%

75%

70%
Dec-06 Dec-07 Dec-08 Dec-09 Dec-10 Jun-06 Jun-07 Jun-08 Jun-09 Jun-10 Sep-06 Sep-07 Sep-08 Sep-09 Sep-10 Jun-11 Mar-07 Mar-08 Mar-09 Mar-10 Mar-11 Sep-11

Source: CBK, Legae Calculations

Page 12 of 22

Fig 12: The regulatory provisioning requirements are fairly tight in our view.

General definition Normal Watch Loans performing in accordance with loan terms Loans exhibit weakness. Examples among other include weakening collateral, deteriorating economic conditions, adverse trend for borrower Loans not adequately protected by current sound net worth and paying capacity of the borrower Loans shows weaknesses in collection in full. The possibility of loss is high Loans are considered uncollectable and continuance recognition as bankable assets is not warranted

Past due by (days) n/m 30 to 90

CBK Regulatory provisions 1% 3%

Substandard

90 to 180

20%

Doubtful

> 180

100%

Lost

> 360

100%

Source: CBK, Legae Securities

Page 13 of 22

2. A look at our universe: Who is swimming naked?

Who has been assuming most of this risk? In a system that could be building credit risk faster than we think, we show on Fig 13, who has been gaining market shares and possibly assuming the greater part of the risks. The biggest market shares loser is Barclays, which was #1 bank in CY05 with a market share of 17% on the system assets. Barclays is now #2, with a market share of 10.3% (CY10). The biggest gainer is Equity, moving from #10 in CY05 and with a market share of 1.9% to #5 in CY10 with a market share of 8% of the systems assets. Equity has gained market shares primarily at the expense of Barclays, ShanChart and Citi. It is however important to note that market share gains do not necessarily imply client acquisition from competitors. The natural growth in the system ensures that even Barclays managed to grow its balance sheet despite losing a vast amount of market share. System assets almost tripled in 5 years; Equity loan book expanded by 14.2x: Looking at the absolute figures, the systems balance sheet has expanded 2.6x between CY05 and CY10. The system loan book has expanded by 2.3x to Kes876bn in CY10 from Kes382bn in CY05. Barclays has assumed the least risk (ignoring loan book quality for now) with its balance sheet expanding only by 1.6x over the same period. Barclays loan book has also grown the lowest at 1.3x to its CY05 value of Kes65.6bn to Kes87.1bn in CY10. Equity assumed the greatest risks (growth) as its balance sheet and loan book have expanded by an exceptionally large 12.5x and 14.2x respectively between CY05 and CY10. This is a strong outperformance to the system growth. Equity is followed by KCB whose balance sheet ballooned from Kes74.3bn in CY05 to Kes251.4bn in CY10. The loan book has also expanded by a colossal 4.5x to Kes148.1bn in CY10. Cooperatives growth approximated system growth on both assets and loans with a 3.0x. Cooperatives total assets increased from Kes51.8bn in CY05 to Kes154.3bn in CY10 whilst the loan book enlarged from Kes29.1bn to Kes86.6bn over the same period. StanChart, like Barclays, lagged system growth, with total assets doubling to Kes142.7bn and the loan book expanding by 1.8x to Kes60.3bn (see Fig 14).

Page 14 of 22

Fig 13: Equity bank has gained enormous market shares mainly at the expense of Barclays, StanChart and Citi...
Institution Institution KCB Barclays Cooperative StanChart Equity Bank CFCStanbic Comm. Bank of Africa Citibank NBK Diamond Trust NIC Total Institution insutitution KCB Barclays Cooperative StanChart Equity Bank CFCStanbic Comm. Bank of Africa NBK NIC Diamond Trust Citibank Total 2005 12.1% 16.7% 8.7% 11.9% 1.8% n/a 5.3% 5.4% 3.4% 2.7% 4.7% 72.6% 2005 12.1% 17.0% 8.4% 11.8% 1.9% n/a 4.8% 5.0% 5.3% 2.6% 3.4% 72.3% Assets Market share 2007 11.8% 16.6% 6.9% 9.6% 5.6% n/a 4.2% 5.0% 4.4% 3.2% 3.3% 70.5% 2009 12.7% 12.2% 8.2% 9.2% 7.1% 7.2% 4.3% 3.8% 3.8% 3.5% 3.3% 75.2% 2010 13.3% 10.3% 9.2% 8.5% 8.0% 6.4% 3.8% 3.7% 3.6% 3.5% 3.3% 73.6% Rank 2005 2 1 4 3 10 n/a 6 5 8 9 7 2007 2 1 4 3 7 n/a 6 5 9 10 8 2009 1 2 3 4 5 6 7 8 9 10 11 2010 1 2 3 4 5 6 7 8 9 10 11 2005 2 1 4 3 10 n/a 4 6 5 9 8 2 1 4 3 5 n/a 8 6 7 10 9 Rank 2007 2009 1 2 4 3 6 5 7 9 8 10 11 2010 1 2 3 4 5 6 7 8 9 10 11

Deposits market share 2007 12.1% 15.4% 7.7% 10.4% 4.4% n/a 4.7% 4.9% 3.5% 3.4% 4.2% 70.7% 2009 13.7% 12.5% 9.1% 8.6% 6.5% 5.6% 4.4% 4.2% 3.7% 3.6% 3.3% 75.2% 2010 13.2% 10.0% 10.0% 8.1% 7.7% 5.9% 4.3% 3.9% 3.7% 3.6% 3.1% 73.5%

Source: CBK, Legae Calculations

Fig 14...and has expanded its balance sheet 12.5x between CY05 and CY10 vs. system average of 2.6x.

Barclays Cooperative Equity KCB Stanchart System

2005 104 522 51 835 11 453 74 338 72 970 636 731

Assets 2010 CAGR 172 415 10.5% 154 340 143 018 251 356 142 746 1 678 112 24.4% 65.7% 27.6% 14.4% 21.4%

Growth x 1.6 3.0 12.5 3.4 2.0 2.6

2005 65 562 29 089 5 524 32 849 34 043 381 544

Loans 2010 CAGR 87 147 5.9% 86 618 78 302 148 113 60 337 876 357 24.4% 69.9% 35.1% 12.1% 18.1%

Growth x 1.3 3.0 14.2 4.5 1.8 2.3

Source: CBK, Company reports, Legae Calculations

Page 15 of 22

...And who is swimming naked? To an extent, the comparison is impaired by the fact that banks report different items in their balance sheet and income statements of accounts. Some report provisions charges, some bad debt charges and other impairment charges, among other variants. Nonetheless, we analyse the credit costs that each bank has pushed through its income statement vs. loan growth to see how this charge has developed vs. loan growth. We note that Cooperatives loan book has significantly outperformed its credit costs growth, with a loan growth of 25% vs. credit costs growth of -13.5%. As the loan book expanded to Kes86.6bn, credit costs declined to Kes799mn from Kes1.4bn. We understand that, like the system, Cooperative has some legacy issues particularly borne in the early part of the past decade, but we do not believe that the current trend is sustainable. It is also out of synch with peers. Our view is irrespective of CBKs Supervision reports that point to considerable improvements in credit screening and monitoring procedures. StanChart credit costs growth has also lagged loan growth, with credit costs growing by 3% while loans expanded by 11%. Barclays and KCBs credit costs growth largely mirror loans and advances growth rate. For Equity, possibly as a reflection of its elevated credit risks, its credit costs soared by 94.3% vs. a loan growth of 48.3% between CY06 and CY10. (see Fig 15). A look at the 3Q11 state of affairs: Looking at 3Q11, we note that StanChart has the best performing loan book with a NPLs ratio of 1.2% (vs. an average of 4.9% for our universe). Barclayss NPL ratio is high at 5.8% (2nd to KCBs 7.9%) chiefly as it does not indicate any suspended interest. However, the coverage ratio for Barclays is also high at 87% due to a high amount of specific provisions, hence a low Net NPLs ratio. We should highlight the worryingly high ratios for KCB (vs. peers), with a NPLs ratio of 7.9% (vs. average of 4.9% for the universe). Provisions are ~50% of the total NPLs amount of Kes12.2bn. As at 3Q11, KCB accounted for ~60% of our universes NPLs vs. a contribution of ~30% to the universe loan book. (see Fig 16).
Fig 15: Cooperative banks credit costs have declined despite a strong growth in loans and advances. Hopefully no surprises in future.

Barclays Cooperative Equity KCB Stanchart System

2006 73 907 28 421 10 930 40 659 35 762 396 149

2010 87 147 86 618 78 302 148 113 60 337 876 357

Loan CAGR 3.4% 25.0% 48.3% 29.5% 11.0% 17.2%

Growth, x 1.2 3.0 7.2 3.6 1.7 2.2

2006 881 1 425 133 735 502 7 672

Credit costs Difference 2010 CAGR Growth, x 1 200 8.0% 1.4 4.7% 0.0% 799 -13.5% 0.6 -38.4% 0.0% 1 905 2 144 447 11 048 94.5% 30.7% -2.9% 9.5% 14.3 2.9 0.9 1.4 46.2% 0.0% 1.2% 0.0% -13.9% 0.0% -7.7%

Source: CBK, Company reports, Legae Calculations

Page 16 of 22

Fig 16: Who was swimming naked by Q311? KCB has the highest NPL ratio

Loans and advances Gross NPLs Suspended interest Total NPLs Loss Provisions Net NPLs Gross NPLs/Loans and advances Suspended interest/Loans and advances Total NPLs/Loans and advances Loss provisions/Loans and advances Net NPLs/Loans and advances Provisions/NPLs

Barclays 98 901 5 699 0 5 699 4 941 758 5.8% 0.0% 5.8% 5.0% 0.8% 86.7%

Cooperative 106 434 5 585 1 056 4 529 3 668 861 5.2% 1.0% 4.3% 3.4% 0.8% 81.0%

Equity 109 367 3 503 542 2 961 1 523 1 438 3.2% 0.5% 2.7% 1.4% 1.3% 51.4%

KCB 174 464 13 834 1 684 12 150 6 254 5 896 7.9% 1.0% 7.0% 3.6% 3.4% 51.5%

StanChart 94 390 1 500 354 1 145 441 704 1.6% 0.4% 1.2% 0.5% 0.7% 38.5%

Source: Company reports, Legae Calculations

Forecasts and recommendations: We show our universes salient assumptions on Fig 17. We underscore the following: We model our loans and advances estimate as a product of the deposits and the LDR. As a result our loans and advances forecasts are directly a result of our deposit growth and LDR estimates. For Barclays, our deposit growth of 5% and 7.5% for FY12 and FY13 respectively are risky given the banks unpredictable strategy. As a result, our growth forecasts could outperform/underperform materially as recent growth rates have been unpredictable. Barclays has historically managed credit risks well, but strategy goes beyond credit risk management and we wonder whether the bank will better execute (and communicate) its strategy than in the past. Generally, we expect banks in our universe to continue to grow EPS, supported by resilient interest spreads and momentum in fee and commission income growth. We do not believe that there is sufficient room for further reductions in credit costs (both in an absolute basis and as ratios) hence improvements in credit costs should have minimal impact to earnings. For FY12 we expect EPS growth of 5.4% for Barclays. Coop (+5.4%); Equity (+10.8%) and KCB (+9.7%) will be affected by relative higher FY11 bases. Our StanChart growth of 25.8% is the highest in our universe, but remains materially lower than consensus by 12%.

Page 17 of 22

Our EPS forecasts vs. consensus: For FY12, we are above consensus on KCB, (+12%), Equity (+4%) and Barclays (+2%). For KCB, we believe investors are underestimating the banks optimal mix between its corporate and retail segment which should bode well for its credit costs development (vs. Equity and Coop). We believe investors could also be underestimating the benefits of its regional subsidiaries and the growing market share in mortgage lending. For Equity bank, our above consensus is also pivoted on regional contribution and further credit improvements in the local market and efficiencies. The primary caveats to our analysis vs. consensus are: a) some of the consensus EPS are averages of a small number of analysts. For example Barclays and StanCharts consensus EPS are based on only 5 analysts; and b) some of the consensus figures could be stale. We reiterate our BUY recommendations on KCB, Equity and Cooperative; Investor sentiment is positive, NSE ALSI up by ~10% YTD: The current market sentiment is on average bullish, with the NSE ALSI up by 9.9% on a YTD basis. Banks have benefited from this positive sentiment, with KCB and Equity share prices increasing by 26.1 and 15.9% respectively. Coop, which we initiated with a SELL recommendation has lost 2.4%, and the valuation risks has reduced to levels we believe provide an attractive risk-return profile. (see Fig 18 - Fig 19). We maintain our BUY recommendations on Coop, Equity and KCB with forecast potential total returns of 28.2%, 27% and 26.8% in that order. However, we believe that in the short-term, Coop may continue to suffer from poor sentiment (perceived poorer franchise, higher credit risks etc). The catalyst would be a stronger delivery in earnings, to include the quality of earnings.
Fig 17: Our salient assumptions
Barclays 2012F Deposit growth Deposits LDR Loans and advances Loan growth Interest income/IEA Interest expense/IBL NIR/Total assets Credit costs/Loans Profit Growth EPS C onsensus EPS Variance 5.0% 130 418 80.0% 104 334 5.3% 12.0% -1.10% 9.3% -0.9% 8 884 10.0% 1.64 1.61 2% 2013F 7.5% 140 199 80.0% 112 159 7.5% 12.0% -1.25% 9.8% -1.0% 9 602 8.1% 1.77 1.79 -1% Coop 2012F 20.0% 182 101 75.0% 136 576 20.0% 8.7% -2.11% 4.7% -1.3% 7 124 5.4% 2.04 2.09 -2% 2013F 17.5% 213 969 75.0% 160 477 17.5% 8.7% -2.20% 4.5% -1.3% 8 007 12.4% 2.29 2.47 -7% Equity 2012F 22.5% 159 909 80.0% 127 927 22.5% 12.6% -1.49% 6.8% -2.3% 11 566 10.8% 3.12 3.00 4% 2013F 20.0% 191 891 80.0% 153 513 20.0% 12.4% -1.48% 7.0% -2.3% 13 383 15.7% 3.61 3.36 8% 2012F 20.0% 311 171 80.0% 248 936 25.3% 10.7% -1.25% 4.9% -1.0% 12 046 9.7% 4.08 3.66 12% KCB 2013F 17.0% 364 070 80.0% 291 256 17.0% 10.5% -1.25% 4.8% -1.0% 13 805 14.6% 4.68 4.12 14% StanChart 2012F 15.0% 132 917 75.0% 99 687 15.0% 8.5% -1.00% 3.3% -1.0% 5 839 25.8% 20.34 23.21 -12% 2013F 15.0% 152 854 75.0% 114 641 15.0% 8.5% -1.25% 3.3% -1.0% 6 768 15.9% 23.57 23.49 0%

Source: Company reports, Legae Calculations

Page 18 of 22

Fig 18: We continue to favour local banks with regional exposure and which are building stronger mortgage lending books although we keep an eye on credit risks

Barclays Current price Price at initiation Loss since Initiation Current rec. Recommendation at initiation FY 12 TP Potential Dividend yield Total potential 13.85 16.7 -17.1% HOLD HOLD 14.57 5.2% 8.3% 13.4%

Coop 11.95 18.25 -34.5% BUY SELL 14.60 22.2% 6.0% 28.2%

Equity 19 24.75 -23.2% BUY BUY 22.88 20.4% 6.6% 27.0%

KCB 21.25 22.25 -4.5% BUY BUY 24.89 17.1% 9.6% 26.8%

StanChart 175 262 -33.2% HOLD HOLD 164.69 -5.9% 8.1% 2.2%

Source: Bloomberg, Legae Calculations

Fig 19: KCB share price has rebounded strongly from 3Q11 lows and has increased by 26% on a YTD basis; Coop is down 2%
1.2 1.1 1.0 0.9 NSE ALSI 0.8 StanChart 0.7 0.6 0.5 9.4% 9.9% Equity 15.9% Barclays Coop Equity KCB StanChart NSE ALSI

YTD, local currency return

KCB 26.1%

Barclays

6.1%

13-Apr-11

15-Jun-11

06-Jul-11

27-Jul-11

07-Sep-11

28-Sep-11

01-Feb-12

22-Feb-12

02-Mar-11

23-Mar-11

19-Oct-11

21-Dec-11

17-Aug-11

04-May-11

25-May-11

Source: Bloomberg, Legae Calculations

09-Nov-11

30-Nov-11

11-Jan-12

Coop -2.4%

-5.0%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

Page 19 of 22

3. A note on KCB FY11 results: Strong performance

FY11 beat our above-consensus: KCBs FY11 EPS of Kes3.72 beat our expectation of Kes3.14 (to include our initial EPS forecast of Kes3.48 which we actually revised downwards in 3Q11) as well as consensus of Kes3.06. The Chairmans statement indicates transformation initiatives that increased revenues as well as efficiency. The turnaround in regional operations that all recorded profitability was also constructive to earnings growth. Legae forecasts vs. Actual Our lower deposit forecast meant lower loans and advances and lower interest income: Our deposit growth of 20% was lower than the 32% growth rate that the bank delivered, taking deposits to Kes259.308bn. As a result, most of the income statement forecasts that depend on deposit growth lagged. We had modelled an 80% LDR vs. the banks 77% LDR, slightly below our expectation but due to a stronger deposit growth the loans and advances was higher than our forecast at Kes198.724 vs. Kes189.095bn. Interest income was Kes27.9bn vs. Legae est. of Kes27.6bn while interest expense was slightly lower at Kes4.62bn vs. our estimate of Kes4.7bn. The resultant net interest income was 2% higher at Kes23.3bn vs. Kes22.9bn. The fee and commission income was stronger as well as Kes9.18bn vs. our est. of Kes7.98bn. As a result total operating income was 7% higher than our expectation at Kes39.3bn vs. 36.7bn. Loan loss was in line at Kes1.89bn vs. our expectation of Kes1.89 but an effective outperformance given our lower loans and advances forecast. Operating expenditure came in 3% higher than our expectation of Kes21.6bn while tax was also higher at Kes4.72bn vs. our forecasts of Kes3.96. The bottom line was higher than we anticipated at Kes10.89bn vs. Kes9.26bn. Assumption changes and our forecasts We grow deposits by 20% in FY12; LDR at 80%: We issued this report before engaging management as the teleconference date was yet to be agreed on. We find it frustrating. We hope in future there will be improved communication especially with foreign analysts/investors. Nonetheless, we provide our forecasts, pivoted primarily on our understanding of KCB and the Kenyan system than management guidance. Fig 20 below shows our leading assumptions. The salient assumptions are : Deposit growth of 20%: We have grown deposit by 20% for FY12 and maintain our initial LDR of 80%; Interest income/interest earnings asset and the costs/asset ratios inferior to history: Our interest income/IEA is inferior to history as we expect interest rates to decline this year. The operating cost/assets ratio is higher than history to accommodate further regional expansion costs; and On average in line with history: On average our assumptions are in line with history although our deposit growth is below the 5year average. Our resultant EPS growth is below history as well.

Page 20 of 22

Valuation: We increase TP to Kes24.89, we maintain our BUY: Our new TP is Kes24.89, primarily due to the increase in the Book value per share, as our Justified PBVR remained constant at 1.4x (rounded off). We maintain our CoE of 19.5% but the average ROE has increased to 22.7%. We forecasts a dividend yield of 9.5% (50% payout ratio) and our total potential return of 26.8%. We maintain our BUY recommendation. Risks to our recommendation/forecasts: 1) Credit costs: Despite improvements in the system, credit risks continue to worry us (as detailed in this report). For KCB, there was a significant improvement in NPLs between 3Q11 and FY11, aided by stronger recoveries. However, the elevated inflation level, higher interest rates in 2H11 and a robust loan growth are unconstructive to bad debts and NPLs formation; 2) macro and politics: macro and political risks, could grow as election draws near; and 3) possible adverse margin development: In FY11, our interest/IEA ratio reduced to 10.7% from 12.2% in FY09 and an average of 11.8% for the past 5 year. Meanwhile, cost of deposits has reduced slightly to 1.6% (from 1.7% in FY10 and vs. an average of 1.6%). Competition seems to be reducing asset yield and increase costs of deposits in the system, notwithstanding the high interest spreads, which is unhelpful to spreads and margins.
Fig 20: Salient assumptions: Below history on deposit growth and earnings growth for FY12 due to FY11 high base; below history on interest/IEA as we expect lower interest rates.
OLD 2007 Salient assumptions Deposit growth rate Loan/Deposit ratio Interest income/Interest earning assets Interest expense/interest earning liabilities Fees and commission income/Total assets Fees and commission expense/Total assets Dividend income/total assets Foreign exchange income/Total assets Other operating income/total assets Bad and doubtful debts expense/loans Other operating expenses/total assets Taxation/PBT EPS n/m 68.1% 11.0% -0.9% 3.8% -0.2% 0.0% 0.7% 0.4% -1.2% -7.6% -29.6% 1.50 34.2% 73.8% 12.9% -1.8% 3.0% -0.1% 0.0% 0.9% 0.2% -1.5% -6.3% -30.3% 2.00 28.7% 75.2% 12.5% -2.1% 3.0% -0.2% 0.0% 0.8% 0.5% -0.6% -8.0% -35.2% 1.84 20.8% 75.2% 12.2% -1.7% 2.8% -0.1% 0.0% 1.1% 0.6% -1.4% -7.4% -26.7% 2.76 31.6% 76.6% 15.0% 80.0% 20.0% 80.0% 10.7% -1.3% 3.3% -0.1% 0.0% 1.1% 0.7% -1.0% -7.8% -29.8% 4.08 10.0% 80.0% 12.0% -1.5% 3.0% -0.1% 0.0% 1.1% 0.8% -1.2% -7.5% -29.1% 4.94 17.0% 80.0% 10.5% -1.3% 3.0% -0.1% 0.0% 1.1% 0.7% -1.0% -7.5% -27.9% 4.68 28.8% 73.8% 11.8% -1.6% 3.1% -0.1% 0.0% 0.9% 0.5% -1.1% -7.2% -29.8% 25.5% 2008 2009 2010 2011 2012F NEW 2012F OLD 2013F NEW Historical

2013F 5-year Av.

10.7% 12.3% -1.6% 2.8% 0.0% 0.0% 1.1% 1.0% -1.0% -6.7% -27.4% 3.72 -2.0% 3.0% -0.1% 0.0% 1.1% 0.7% -1.0% -7.5% -30.6% 3.58

Source: Company reports, Legae Calculations

Page 21 of 22

Disclosure & Disclaimer

Legae Securities (Pty) Ltd Member of the JSE Securities Exchange 1st Floor, Building B, Riviera Road Office Park, 6-10 Riviera Road, Houghton, Johannesburg, South Africa P.O Box 10564, Johannesburg, 2000, South Africa Tel +27 11 551 3601, Fax +27 11 551 3635 Web: www.legae.co.za, email: research@legae.co.za

Analyst Certification and Disclaimer I/we the author (s) hereby certify that the views as expressed in this document are an accurate of my/our personal views on the stock or sector as covered and reported on by myself/each of us herein. I/we furthermore certify that no part of my/our compensation was, is or will be related, directly or indirectly, to the specific recommendations or views as expressed in this document This report has been issued by Legae Securities (Pty) Limited. It may not be reproduced or further distributed or published, in whole or in part, for any purposes. Legae Securities (Pty) Ltd has based this document on information obtained from sources it believes to be reliable but which it has not independently verified; Legae Securities (Pty) Limited makes no guarantee, representation or warranty and accepts no responsibility or liability as to its accuracy or completeness. Expressions of opinion herein are those of the author only and are subject to change without notice. This document is not and should not be construed as an offer or the solicitation of an offer to purchase or subscribe or sell any investment. Important Disclosure This disclosure outlines current conflicts that may unknowingly affect the objectivity of the analyst(s) with respect to the stock under analysis in this report. The analyst(s) do not own any shares in the company under analysis.