October 28, 2020 Herd Immunity: Refining Our Model Based On Latest Developments

(1)

EQ

U

IT

Y

RE

SE

A

RC

Leonid Timashev, Ph.D. (Associate) (212) 437-9931 [email protected] Gregory Renza, M.D. (Analyst) (212) 858-7065 [email protected] October 28, 2020

Herd Immunity: Refining Our Model Based On Latest

Developments

See Potential For Herd Immunity By 1Q23, Best Case 2Q21; With

Vaccine By 3Q21

We

recently analyzed

the potential of reaching COVID-19 herd immunity through either

widespread infections and/or the common use of a vaccine. As we continue to await

vaccine data, and as case rates continue to rise to record levels, hospitalizations steadily

increase, and early signs of mortality also begin to inflect upwards since our last note

on 10/15, we take the opportunity to update our thinking and further refine our model

assumptions, shaped also by emerging scientific data.

We have also constructed an

interactive scenario analysis to explore base, best, and worst cases using a variety of

customizable inputs.

Our updated base case analysis indicates that without a vaccine,

the U.S. should achieve sufficient effective immunity to meaningfully slow transmission

and restore societal 'normalcy' by 1Q23, resulting in 660K COVID-19 related deaths (low

145K, high 1M). Assuming a higher proportion of individuals already immune and a lower

infection-related mortality rate, both of which may be very reasonable assumptions based

on evolving evidence, lowered spread would occur by 2H22 at the cost of 245K U.S. lives.

Should an effective vaccine become available over the next few months (first interim data

are imminent), our analysis indicates a meaningful slowing of transmission by 3Q21, with

up to an additional 140-340K COVID-related deaths occurring between now and then.

Herd immunity model with inputs and sensitivities available here.

Latest evolving understandings and trends:

Rising case rates: The 7-day moving average for positive COVID-19 cases in the U.S. has now reached record highs at 72K/day (Exhibit 1), and we are increasing our infection rate projections for the coming months accordingly to reflect these trends.

Testing up, but so are hospitalizations: In the early-summer wave the uptick in hospitalizations lagged the uptick in cases by about 2wks, and we see a similar effect in the current wave-- indicating that the spike in positive cases is reflecting a true increase in COVID-19 infections rather than solely due to better detection; after adjusting for that 2-week lag, it appears that the % increase in hospitalizations is tracking at least on par with the case uptick. However, while the ratio of total COVID-related hospitalizations as a function of new positive cases 14d prior has remained relatively constant (0.8) since August, suggestive of some equilibrium, it remains substantially lower than during the summer wave, and the hospitalization rate is far lower (half) than what was seen in the spring – indicating that there could be some contribution from detection of milder/asymptomatic cases and that hospitalization numbers may not substantially surpass prior waves despite higher positive tests (though this may also be due in part to more stringent hospitalization criteria). This suggests much better manageability vs. early in the pandemic, and shapes our assumptions particularly in our “optimistic base case 2.0.”

Death rate steady since mid-July; remains far lower than early in pandemic:

(cont. on next page)

(2)

(cont. from front page)

Death rates (continued): Since mid-July, the rate of reported COVID-related deaths relative to new positive cases 3wks prior (~15d diagnosis to death + ~7d median time to report death) has continued to hover between 1.6-2.0%, indicating mortality rates are generally steady and remain far lower than at the beginning of the pandemic.Accounting for the lag period, though, it does not appear that the death rates from COVID-19 will be substantially lower during this wave than in the mid-summer wave. With our estimate that there are actually 2.3x the number of daily infections vs. positive tests (discussed later), this case fatality rate corresponds roughly to our 0.7% estimated infection fatality rate. Despite some recent setbacks, including more mixed data for GILD’s remdesivir and failure of LLY’s bamlanivimab in hospitalized patients, we continue to model, optimistically, that with improving care and greater therapeutic options, death rates should still decline over time. With cases distributed more evenly throughout the country vs. early in the pandemic, as well as greater preparedness, we see less risk of regional healthcare systems being completely overrun (we estimate 78K beds could be occupied at peak vs. an estimated 302K available beds), though we will continue to monitor this given its potential to raise mortality rates.

Exhibit 1:Hospitalizations and deaths have risen consistent with positive cases, albeit with a lag period

Source: RBC Capital Markets estimates; COVID Tracking Project

Duration of immunity: Studies have shown immunity to other coronaviruses can last as long as 3-5 years, reinfections with COVID-19 have been very rare, and protective immunity with COVID-19 has been demonstrated to be at least several months. However, a large new study just published out of Imperial College London showed the proportion of people testing positive for COVID-19 antibodies declined by 27% over 3 months, and loss of antibodies was much greater in individuals with asymptomatic infection. There are many unknowns, including the degree that other forms of immunity like T-cells may still confer protection, and we note that undetectable levels of antibodies does not necessarily mean there are no memory B cells readily available to differentiate into plasma cells and prevent a second infection on stimulation with antigen, but nonetheless it does raise some questions about the extent and duration in which asymptomatic patients can contribute to herd immunity. To better account for this, we are introducing a very modest adjustment factor into our model to slightly reduce the proportion of durably immune patients by 0.3% each month in our base case.

(3)

Warm weather and transmission: Emerging evidence continues to suggest summer, and warmer weather, do not have a meaningful impact on the effective reproduction number for COVID-19, with a recent pre-print from Iran showing nearly no effect to a stable 2.5-2.6 Rt throughout a 10 degree Celsius range. As such, we maintain our transmission estimates constant for the Northern Hemisphere winter season.

Pre-existing T cells may protect against more severe disease, though not necessarily against transmission: Our recent review of relevant literature suggests pre-existing T cells may lessen severity of COVID-19 infection. However, they may have limited effect on transmission rates, and consequently, on herd immunity thresholds. We model a 2% reduction in the herd immunity threshold to account for the possibility of pre-existing immunity.

Increasing characterization of longer-term symptoms: A recent pre-print analyzed self-reported short- and long-term symptoms in a mostly mild COVID-19 cohort vs two control cohorts and found that 43% of COVID-19 patients had symptoms lasting more than 30d vs just 9% in the general untested population. Further, 24% of these mild COVID-19 patients still reported symptoms after 90d. It is difficult to know the degree to which milder patients may have any lingering effects that extend beyond weeks/months that could impact their quality of life more sustainably. Overall, we believe longer-term symptoms remain an important consideration, as many models look solely at hospitalizations and mortality but not at potential long-term morbidity.

Rationale for our assumptions:

Intrinsic Rt: The best estimate for the intrinsic reproductive number is currently 2.5 per the CDC, and we assume a best and worst case of 2.0 and 3.0, respectively.

Initial immunity: We see some debate on this variable; however, we continue to believe the totality of evidence to date on SARS-CoV-2 seroprevalence implies an average U.S. immunity rate of around 10% (15% in our base 2.0 case). We assume 14% and 8% for our best and worst case, respectively, for the proportion of the population previously infected with COVID-19.

Bar for herd immunity: Our estimated bar for herd immunity is roughly 43% of the population, which is based on a 60% threshold from the Rt of 2.5, less 15% to account for COVID-19's heterogeneous spread and impact of social behaviors, and 2% for pre-existing complete immunity. This threshold is 33% in our best case and 50% in our worst case scenarios. Our model continues to assume a balance of openings/closures and social distancing as it stands today; greater lockdowns may lower the effective herd immunity threshold, while reduced distancing and isolation behaviors would increase it.

Neutralizing cross-reactivity: Another topic of debate with data limitations is whether cross-reactivity from pre-COVID-19 coronaviruses can be neutralizing and/or protective. Though difficult to precisely estimate, we assume 2% of the population may have complete protection (equivalent to 4% with half-protection, etc.), with a best and worst case of 5% and 0.5%, respectively.

Reinfection: With fewer than a handful of cases of reinfections worldwide, we see the base case rate at 0.005% and the best case at 0%, with a worst case scenario of 0.01%, and these do not factor significantly in our modeled outcomes.

Initial detected / undetected ratio: We maintain our base case ratio of 2.3 (meaning there are actually 2.3x as many infections as positive tests), after back-calculating from multiple sources. However, we acknowledge this is a key variable, and if there are multiples more of the population getting infected - perhaps asymptomatically - than are showing up in testing, then

(4)

the path towards achieving herd immunity could be more rapid. Our best-case ratio of 4x would account for this effect and could certainly be realistic, in our view.

Vaccine efficacy: Putting a finer point on vaccine impact to herd immunity, we estimate base case vaccine efficacy to be in the 65% range, with a best case of 90% (similar to non-flu vaccines), and a worst case as the minimum efficacy bar set by the FDA at 50%. We are also nuancing the vaccine impact based on the potential timelines for availability, likely requirement for two doses, and several weeks to development of immunity after vaccination.

Infection fatality ratio (IFR): We base our 0.70% IFR (0.30% in our base 2.0 case) on estimates provided by several sources including the CDC and Johns Hopkins University, as well as our back-calculations from the steady lag-corrected U.S. death rate trends. Our best-case IFR rate of 0.30% would imply many more infections than being detected by tests; recent published statistical work correcting for demographics suggest this could be the case. Additionally, a new publication in Nature this week suggests the more predominant D614G mutation, while causing increased upper respiratory tract titers (and likely associated increased transmissibility), may be more susceptible to neutralization – perhaps contributing to improvements in mortality vs. very early in the pandemic.

IFR decay rate: Our IFR decay rate of 2% is based on assumptions surrounding improvements in care and availability of therapeutics. We model best- and worst-case rates of 3% and 0%.

Rapidity of herd immunity: We are also nuancing our model to better account for an expected decline in cases/transmission to begin rapidly (-70% per month) after achieving the herd immunity threshold, but cases not necessarily disappearing immediately.

Exhibit 2:Without a vaccine, we est. herd immunity could potentially be achieved in the U.S. around early 2023

Source: RBC Capital Markets estimates; based on data from Meteus et al Science (2020), Anand et al Lancet (2020), Poland et al Lancet (2020), Cirulli et al medRxiv (2020, pre-print), Plante et al Nature (2020), REACT-2 (gov.uk), IHME

0% 10% 20% 30% 40% 50% 60% 0 50 100 150 200 250 300 350 400 Im m u n e , e n d o f m o n th (%) P o p u lat ion (00 0 s)

Daily Infections Per Month Monthly Deaths % Immune

(5)

Exhibit 3:With a vaccine, we est. herd immunity could potentially be achieved in the U.S. by late-summer 2021

Exhibit 4:Table of scenario outcomes (including an “optimistic base 2.0” case – see footnote)

Note: Base 2.0 scenario is the same as Base scenario except starts at a 15% (vs 10%) current infection rate and lower 0.3% (vs 0.7%) IFR

0% 10% 20% 30% 40% 50% 60% 0 50 100 150 200 250 300 350 400 Im m u n e , e n d o f m o n th (%) P o p u lat ion (00 0 s)

Daily Infections Per Month Monthly Deaths % Immune

Worst

Base

Base 2.0

Best

Without a vaccine

Herd immunity reached

Never

Jan-'23

Aug-'22

Apr-'21

Total COVID-19 deaths (000s)

1,032

657

248

145 Total excess deaths (000s)

1,445

919

347

203 With a vaccine

Herd immunity reached

Never

Sep-'21

Aug-'21

Mar-'21

Total COVID-19 deaths (000s)

904

339

137

130 Total excess deaths (000s)

1,266

474

192

181 Difference (with vs without vaccine)

Herd immunity reached

Never

15 mo

12 mo

1 mo

Total COVID-19 deaths (000s)

128

318

111

16 Total excess deaths (000s)

179

445

155

22

(6)

Exhibit 5:Key Upcoming Clinical Trial Readouts for COVID-19

Source: RBC Capital Markets estimates; company reports, clinicaltrials.gov

Drug Company Stage Description Design Anticipated Readout NCT

COVID-19 treatments: small molecules

Galidesivir BCRX ph.Ib

Direct anti-viral blocking viral RdRp, IV administration.

66 moderate to severe pts across part A (dose ranging) & B (efficacy),

randomized double blind

4Q20 NCT03891420

Famotidine Generic Approved in alt. indication

H2 antagonist, MoA in COVID-19 unknown

1170 moderate to severe pts, randomized double blind, Fam + HCQ

vs HCQ alone

4Q20 NCT04370262

Ruxolitinib INCY/NOV Approved in

alt. indication JAK1/2 inhibitor

402 severe pts, randomized, double

blind Oct-'20 NCT04362137 Emtricitabine / Tenofovir GILD/ Generic Approved in alt. indication

HIV combo adenosine and aytosine analogues block

viral polymerase

4000 pts, randomized, double blind for

prophylaxis 4Q20 NCT04334928

Nitazoxanide Romark Approved in alt. indication

Broad spectrum anti-parasitic with anti-viral

properties

800 pts, randomized, double blind for

prophylaxis 2H20 NCT04359680

Molnupiravir (EIDD-2801)

MRK /

Ridgeback ph.II/III

Direct anti-viral blocking viral RdRp, oral

administration.

1300 pt trial of non-hospitalized COVID-19 patients, 1450 pt trial of

hospitalized

1H21 NCT04575597 NCT04575584

COVID-19 treatments: antibodies

REGN-COV2 REGN ph.III

Targets spike protein; two antibody cocktail ('10933

and '10987)

2000 asymptomatic healthy adults in

randomized, DB, pbo-controlled study end-'20/early-'21 NCT04452318

LY-CoV555 LLY ph.II Targets spike protein 800 mild/mod pts in randomized,

double-blinded, pbo-controlled study Jan-'21 NCT04427501 VIR-7831 VIR ph.II/III Targets RBD 1360 non-hospitalized, double-blinded,

pbo-controlled study Jan-'21 NCT04545060 AZD7442 AZN ph.I Two antibody cocktail

(AZD8895 and AZD1061) 48 health volunteers 4Q20 NCT04507256

COVID-19 vaccines

mRNA-1273 MRNA ph.II mRNA delivered by LNP

600 HV; two dose levels; two rounds of dosing 28d apart; two age groups

(18-54; 55+)

Late-'20 NCT04405076

CVnCoV vaccine GSK/CureVac ph.I mRNA delivered by LNP

168 HV; three dose levels; two rounds of dosing with 28d apart; one age

group (18-60)

Late-'20 NCT04449276

SCB-2019 Clover/GSK/D

VAX ph.I

Protein subunit with 2 types of adjuvants or

without adjuvant

150 HV; 3 dose levels; two rounds of dosing 21d apart; 2 age groups

(18-54; 55-75)

End-20 NCT04405908

AZD1222 AZN/Oxford ph.II/III (UK) ChAd vector-based

12,330 HV; one dose level; two dosing schedule (single or booster at wk4-6); five age groups (18+; 18-55;

5-12; 56-69; 70+)

Late-'20 NCT04400838

mRNA-1273 MRNA ph.III mRNA delivered by LNP

30,000 HV; one dose level - 100µg; two rounds of dosing 28d apart; one

age group (18+)

Nov-'20 NCT04470427

BNT162b2 PFE/BNTX ph.II/III mRNA encoding for Spike protein delivered by LNP

up to 30,000 HV; one dose level - 30µg; two rounds of dosing 21d apart;

two age groups (18-55; 65-85)

Late-'20 NCT04368728

SARS-CoV-2

vaccine SNY/GSK ph.I/II

Protein subunit with 2 types of adjuvants or

without adjuvant

440 HV; two formulations with two adjuvants; two dosing schedule (single

or booster 21d apart); two age groups (18-49; 50+)

Nov-'20 NCT04537208

AZD1222 AZN/Oxford ph.III mRNA delivered by LNP

30,000 HV; one dose level - 5*10^10 vp; two rounds of dosing 28d apart;

one age group (18+)

Dec-'20 NCT04516746

JNJ-78436735 JNJ ph.III Ad26 vector-based 60,000 HV; one dose level; one

(7)

Exhibit 6:Model Assumptions Without Vaccine

Vaccines? N

Case 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 26 38 50 62

Above threshold? FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE

HERD IMMUNITY Oct-20 Nov-20 Dec-20 Jan-21 Feb-21 Mar-21 Apr-21 May-21 Jun-21 Jul-21 Aug-21 Sep-21 Oct-21 Nov-21 Dec-21 Dec-22 Dec-23 Dec-24 Dec-25

U.S. population (M) 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330

SARS-CoV-2 Intrinsic Rt 2.5

Intrinsic threshold needed 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60%

Behavioral modifications 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15%

Effective threshold needed 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45%

Needed to reach effective threshold (M) 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149

Immune, beginning of month (%) 12% 13% 14% 16% 18% 19% 21% 22% 23% 25% 26% 27% 28% 30% 31% 44% 45% 43% 41%

Neutralizing Cross-reactivity 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2%

Vaccinated 0% 0% 1% 3% 7% 12% 17% 22% 27% 32% 37% 42% 47% 52% 52% 52% 52% 52% 52%

Newly Infected 1% 2% 2% 2% 2% 1% 1% 1% 1% 1% 1% 1% 1% 1% 1% 1% 0% 0% 0%

Reinfection 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% Immunity Loss 0.04% 0.04% 0.04% 0.05% 0.05% 0.06% 0.06% 0.07% 0.07% 0.07% 0.08% 0.08% 0.08% 0.09% 0.09% 0.13% 0.13% 0.13% 0.12%

Immune, end of month (%) 13% 14% 16% 18% 19% 21% 22% 23% 25% 26% 27% 28% 30% 31% 32% 45% 44% 43% 41%

Immune, end of month (M) 42 46 52 58 64 69 73 77 81 85 90 93 97 101 105 147 147 142 136

Daily reported rate of COVID infections 60.0 90.0 95.0 85.0 75.0 65.0 64.4 63.7 63.1 62.4 61.8 61.2 60.6 60.0 59.4 52.6 0.0 0.0 0.0 Daily rate of COVID infections, proj (000s) 138 207 219 196 173 150 148 147 145 144 142 141 139 138 137 121 0 0 0

Monthly COVID infections, proj (000s) 4,140 6,210 6,555 5,865 5,175 4,485 4,440 4,396 4,352 4,308 4,265 4,223 4,180 4,139 4,097 3,632 0 0 0 IFR 0.70% 0.69% 0.67% 0.66% 0.65% 0.63% 0.62% 0.61% 0.60% 0.58% 0.57% 0.56% 0.55% 0.54% 0.53% 0.41% 0.32% 0.25% 0.20%

Monthly Deaths, Directly COVID-19 Related 29 43 44 39 33 28 28 27 26 25 24 24 23 22 22 15 0 0 0

Undercount 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40%

Monthly Deaths, Excess (000s) 41 60 62 54 47 40 39 37 36 35 34 33 32 31 30 21 0 0 0

Total COVID-19 deaths (000s) 657

(8)

Exhibit 7:Model Assumptions With Vaccine

Vaccines? Y

Case 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 26 38 50 62

Above threshold? FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE

HERD IMMUNITY Oct-20 Nov-20 Dec-20 Jan-21 Feb-21 Mar-21 Apr-21 May-21 Jun-21 Jul-21 Aug-21 Sep-21 Oct-21 Nov-21 Dec-21 Dec-22 Dec-23 Dec-24 Dec-25

U.S. population (M) 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330 330

SARS-CoV-2 Intrinsic Rt 2.5

Intrinsic threshold needed 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60%

Behavioral modifications 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15% 15%

Effective threshold needed 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45% 45%

Needed to reach effective threshold (M) 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149 149

Immune, beginning of month (%) 12% 13% 14% 16% 18% 20% 23% 27% 31% 35% 40% 44% 48% 52% 55% 59% 57% 55% 53%

Neutralizing Cross-reactivity 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2%

Vaccinated 0% 0% 1% 3% 7% 12% 17% 22% 27% 32% 37% 42% 47% 52% 52% 52% 52% 52% 52%

Newly Infected 1% 2% 2% 2% 2% 1% 1% 1% 1% 1% 1% 0% 0% 0% 0% 0% 0% 0% 0%

Reinfection 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% 0.01% Immunity Loss 0.04% 0.04% 0.04% 0.05% 0.05% 0.06% 0.07% 0.08% 0.09% 0.11% 0.12% 0.13% 0.14% 0.15% 0.16% 0.18% 0.17% 0.16% 0.16%

Immune, end of month (%) 13% 14% 16% 18% 20% 23% 27% 31% 35% 40% 44% 48% 52% 55% 58% 59% 57% 55% 53%

Immune, end of month (M) 42 46 52 58 65 75 88 103 117 131 145 159 170 181 191 195 188 181 174

Daily reported rate of COVID infections 60.0 90.0 95.0 85.0 75.0 65.0 64.4 63.7 63.1 62.4 61.8 18.4 5.5 1.6 0.5 0.0 0.0 0.0 0.0

Daily rate of COVID infections, proj (000s) 138 207 219 196 172 147 141 135 129 123 117 33 10 3 1 0 0 0 0

Monthly COVID infections, proj (000s) 4,140 6,210 6,555 5,865 5,158 4,398 4,238 4,053 3,871 3,692 3,517 1,003 286 81 23 0 0 0 0 IFR 0.70% 0.69% 0.67% 0.66% 0.65% 0.63% 0.62% 0.61% 0.60% 0.58% 0.57% 0.56% 0.55% 0.54% 0.53% 0.41% 0.32% 0.25% 0.20%

Monthly Deaths, Directly COVID-19 Related 29 43 44 39 33 28 26 25 23 22 20 6 2 0 0 0 0 0 0

Undercount 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40%

Monthly Deaths, Excess (000s) 41 60 62 54 47 39 37 34 32 30 28 8 2 1 0 0 0 0 0

Total COVID-19 deaths (000s) 339

(9)

Required disclosures

Conflicts disclosures

The analyst(s) responsible for preparing this research report received compensation that is based upon various factors, including total revenues of the member companies of RBC Capital Markets and its affiliates, a portion of which are or have been generated by investment banking activities of the member companies of RBC Capital Markets and its affiliates.

Distribution of ratings

For the purpose of ratings distributions, regulatory rules require member firms to assign ratings to one of three rating categories -Buy, Hold/Neutral, or Sell - regardless of a firm's own rating categories. Although RBC Capital Markets' ratings of Outperform (O), Sector Perform (SP), and Underperform (U) most closely correspond to Buy, Hold/Neutral and Sell, respectively, the meanings are not the same because our ratings are determined on a relative basis.

Distribution of ratings RBC Capital Markets, Equity Research

As of 30-Sep-2020

Investment Banking Serv./Past 12 Mos.

Rating Count Percent Count Percent

BUY [Outperform] 788 52.96 248 31.47

HOLD [Sector Perform] 619 41.60 135 21.81

SELL [Underperform] 81 5.44 11 13.58

Conflicts policy

RBC Capital Markets Policy for Managing Conflicts of Interest in Relation to Investment Research is available from us on request. To access our current policy, clients should refer to

https://www.rbccm.com/global/file-414164.pdf

or send a request to RBC Capital Markets Research Publishing, P.O. Box 50, 200 Bay Street, Royal Bank Plaza, 29th Floor, South Tower, Toronto, Ontario M5J 2W7. We reserve the right to amend or supplement this policy at any time.

Dissemination of research and short-term trade ideas

RBC Capital Markets endeavors to make all reasonable efforts to provide research simultaneously to all eligible clients, having regard to local time zones in overseas jurisdictions. RBC Capital Markets' equity research is posted to our proprietary website to ensure eligible clients receive coverage initiations and changes in ratings, targets and opinions in a timely manner. Additional distribution may be done by the sales personnel via email, fax, or other electronic means, or regular mail. Clients may also receive our research via third party vendors. RBC Capital Markets also provides eligible clients with access to SPARC on the Firms proprietary INSIGHT website, via email and via third-party vendors. SPARC contains market color and commentary regarding subject companies on which the Firm currently provides equity research coverage. Research Analysts may, from time to time, include short-term trade ideas in research reports and / or in SPARC. A short-term trade idea offers a short-term view on how a security may trade, based on market and trading events, and the resulting trading opportunity that may be available. A short-term trade idea may differ from the price targets and recommendations in our published research reports reflecting the research analyst's views of the longer-term (one year) prospects of the subject company, as a result of the differing time horizons, methodologies and/or other factors. Thus, it is possible that a subject company's common equity that is considered a long-term 'Sector Perform' or even an 'Underperform' might present a short-term buying opportunity as a result of temporary selling pressure in the market; conversely, a subject company's common equity rated a long-term 'Outperform' could be considered susceptible to a short-term downward price correction. Short-term trade ideas are not ratings, nor are they part of any ratings system, and the firm generally does not intend, nor undertakes any obligation, to maintain or update short-term trade ideas. Short-term trade ideas may not be suitable for all investors and have not been tailored to individual investor circumstances and objectives, and investors should make their own independent decisions regarding any securities or strategies discussed herein. Please contact your investment advisor or institutional salesperson for more information regarding RBC Capital Markets' research.

For a list of all recommendations on the company that were disseminated during the prior 12-month period, please click on the following link: https://rbcnew.bluematrix.com/sellside/MAR.action

(10)

The 12 month history of SPARCs can be viewed at https://www.rbcinsightresearch.com.

Analyst certification

All of the views expressed in this report accurately reflect the personal views of the responsible analyst(s) about any and all of the subject securities or issuers. No part of the compensation of the responsible analyst(s) named herein is, or will be, directly or indirectly, related to the specific recommendations or views expressed by the responsible analyst(s) in this report.

Third-party-disclaimers

The Global Industry Classification Standard ("GICS”) was developed by and is the exclusive property and a service mark of MSCI Inc. (“MSCI”) and Standard & Poor's Financial Services LLC (“S&P”) and is licensed for use by RBC. Neither MSCI, S&P, nor any other party involved in making or compiling the GICS or any GICS classifications makes any express or implied warranties or representations with respect to such standard or classification (or the results to be obtained by the use thereof), and all such parties hereby expressly disclaim all warranties of originality, accuracy, completeness, merchantability and fitness for a particular purpose with respect to any of such standard or classification. Without limiting any of the foregoing, in no event shall MSCI, S&P, any of their affiliates or any third party involved in making or compiling the GICS or any GICS classifications have any liability for any direct, indirect, special, punitive, consequential or any other damages (including lost profits) even if notified of the possibility of such damages.

RBC Capital Markets disclaims all warranties of originality, accuracy, completeness, merchantability or fitness for a particular purpose with respect to any statements made to the media or via social media that are in turn quoted in this report, or otherwise reproduced graphically for informational purposes.

References herein to "LIBOR", "LIBO Rate", "L" or other LIBOR abbreviations means the London interbank offered rate as administered by ICE Benchmark Administration (or any other person that takes over the administration of such rate).

Disclaimer

RBC Capital Markets is the business name used by certain branches and subsidiaries of the Royal Bank of Canada, including RBC Dominion Securities Inc., RBC Capital Markets, LLC, RBC Europe Limited, Royal Bank of Canada, Hong Kong Branch and Royal Bank of Canada, Sydney Branch. The information contained in this report has been compiled by RBC Capital Markets from sources believed to be reliable, but no representation or warranty, express or implied, is made by Royal Bank of Canada, RBC Capital Markets, its affiliates or any other person as to its accuracy, completeness or correctness. All opinions and estimates contained in this report constitute RBC Capital Markets' judgement as of the date of this report, are subject to change without notice and are provided in good faith but without legal responsibility. Nothing in this report constitutes legal, accounting or tax advice or individually tailored investment advice. This material is prepared for general circulation to clients and has been prepared without regard to the individual financial circumstances and objectives of persons who receive it. The investments or services contained in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about the suitability of such investments or services. This report is not an offer to sell or a solicitation of an offer to buy any securities. Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur. RBC Capital Markets research analyst compensation is based in part on the overall profitability of RBC Capital Markets, which includes profits attributable to investment banking revenues. Every province in Canada, state in the U.S., and most countries throughout the world have their own laws regulating the types of securities and other investment products which may be offered to their residents, as well as the process for doing so. As a result, the securities discussed in this report may not be eligible for sale in some jurisdictions. RBC Capital Markets may be restricted from publishing research reports, from time to time, due to regulatory restrictions and/ or internal compliance policies. If this is the case, the latest published research reports available to clients may not reflect recent material changes in the applicable industry and/or applicable subject companies. RBC Capital Markets research reports are current only as of the date set forth on the research reports. This report is not, and under no circumstances should be construed as, a solicitation to act as securities broker or dealer in any jurisdiction by any person or company that is not legally permitted to carry on the business of a securities broker or dealer in that jurisdiction. To the full extent permitted by law neither RBC Capital Markets nor any of its affiliates, nor any other person, accepts any liability whatsoever for any direct, indirect or consequential loss arising from, or in connection with, any use of this report or the information contained herein. No matter contained in this document may be reproduced or copied by any means without the prior written consent of RBC Capital Markets in each instance.

Additional information is available on request.

To U.S. Residents:

This publication has been approved by RBC Capital Markets, LLC (member FINRA, NYSE, SIPC), which is a U.S. registered broker-dealer and which accepts responsibility for this report and its dissemination in the United States. Any U.S. recipient of this report that is not a registered broker-dealer or a bank acting in a broker or dealer capacity and that wishes further information regarding, or to effect any transaction in, any of the securities discussed in this report, should contact and place orders with RBC Capital Markets, LLC.

To Canadian Residents:

This publication has been approved by RBC Dominion Securities Inc.(member IIROC). Any Canadian recipient of this report that is not a Designated Institution in Ontario, an Accredited Investor in British Columbia or Alberta or a Sophisticated Purchaser in Quebec (or similar permitted purchaser in any other province) and that wishes further information regarding, or to effect any transaction in, any of the securities discussed in this report should contact and place orders with RBC Dominion Securities Inc., which, without in any way limiting the foregoing, accepts responsibility for this report and its dissemination in Canada.

To U.K. Residents:

This publication has been approved by RBC Europe Limited ('RBCEL') which is authorized by the Prudential Regulation Authority and regulated by the Financial Conduct Authority ('FCA') and the Prudential Regulation Authority, in connection with its distribution in the United Kingdom. This material is not for general distribution in the United Kingdom to retail clients, as defined under the rules of the FCA. RBCEL accepts responsibility for this report and its dissemination in the United Kingdom.

(11)

This material has been distributed in Australia by Royal Bank of Canada, Sydney Branch (ABN 86 076 940 880, AFSL No. 246521). This material has been prepared for general circulation and does not take into account the objectives, financial situation or needs of any recipient. Accordingly, any recipient should, before acting on this material, consider the appropriateness of this material having regard to their objectives, financial situation and needs. If this material relates to the acquisition or possible acquisition of a particular financial product, a recipient in Australia should obtain any relevant disclosure document prepared in respect of that product and consider that document before making any decision about whether to acquire the product. This research report is not for retail investors as defined in section 761G of the Corporations Act.

To Hong Kong Residents:

This publication is distributed in Hong Kong by Royal Bank of Canada, Hong Kong Branch, which is regulated by the Hong Kong Monetary Authority and the Securities and Futures Commission (SFC) in Hong Kong, RBC Investment Services (Asia) Limited and RBC Global Asset Management (Asia) Limited, both entities are regulated by the SFC. This material is not for general distribution in Hong Kong to persons who are not professional investors (as defined in the Securities and Futures Ordinance of Hong Kong (Cap. 571) and any rules made thereunder.

To Singapore Residents:

This publication is distributed in Singapore by the Royal Bank of Canada, Singapore Branch, a registered entity licensed by the Monetary Authority of Singapore. This material has been prepared for general circulation and does not take into account the objectives, financial situation, or needs of any recipient. You are advised to seek independent advice from a financial adviser before purchasing any product. If you do not obtain independent advice, you should consider whether the product is suitable for you. Past performance is not indicative of future performance. If you have any questions related to this publication, please contact the Royal Bank of Canada, Singapore Branch. Royal Bank of Canada, Singapore Branch accepts responsibility for this report and its dissemination in Singapore.

To Japanese Residents:

Unless otherwise exempted by Japanese law, this publication is distributed in Japan by or through RBC Capital Markets (Japan) Ltd. which is a Financial Instruments Firm registered with the Kanto Local Financial Bureau (Registered number 203) and a member of the Japan Securities Dealers Association (JSDA) and the Financial Futures Association of Japan (FFAJ).