GPT Group. Lower for longer to low for longer A$4.82 AUSTRALIA

Transcription

1 AUSTRALIA GPT AU Price (at 05:11, 14 Feb 2017 GMT) Neutral A$4.82 Valuation - NAV A$ month target A$ month TSR % Volatility Index Low GICS sector Real Estate Market cap A$m 8, day avg turnover A$m 22.9 Number shares on issue m 1,798 Investment fundamentals Year end 31 Dec 2016A 2017E 2018E 2019E Revenue m EBIT m Reported profit m 1, Adjusted profit m Gross cashflow m CFPS CFPS growth % PGCFPS x EPS adj EPS adj growth % PER adj x Total DPS Total DPS growth % Total div yield % FFO yield % AFFO yield % Franking % ROA % ROE % EV/EBITDA x Net debt/equity % P/BV x GPT AU vs ASX 200 Prop, & rec history Note: Recommendation timeline - if not a continuous line, then there was no Macquarie coverage at the time or there was an embargo period. Source: FactSet, Macquarie Research, February 2017 (all figures in AUD unless noted) 15 February 2017 Macquarie Securities (Australia) Limited Lower for longer to low for longer Event GPT reported FY16 statutory NPAT of $1,152.7m. FFO of 29.9cps was directly in line with our 29.9cps forecast (drivers discussed overleaf). Impact FY17 guidance also in line with expectations...growth rate to slow back to low single digit as expected. Consistent with our prior forecast for FY17 FFO growth of 1.7%, GPT guided to 2.0% growth in FY17. With GPT again targeting development profits of ~1-2% of FFO, we note earnings would be flat excluding this activity. Further out (FY20+), we note it gets harder with the base hedge rate increasing to 4.08% vs 3.12% currently (~5.7% all-in using current margins) placing a drag on earnings (MRE FY20: +0.0%). Interestingly, GPT guided to DPS growth of 5.0% in FY17 (MRE prior: +4.4%). Whilst GPT highlighted its DPS is AFFO covered, the DPS is short of AFFO treating the MLC façade restoration as maintenance activity ($18m spend remaining). Despite this, incentives are expected to fall due to the strong Sydney office market, aiding FCF coverage of DPS. Lots of capacity but limited opportunities answer at the moment is development. Factoring strong revaluations partially offset by recent GWSCF/GWOF redeployment, GPT had 23.7% gearing at December (24.4% at June). With the group recently extending its on market share buyback, we believe the group will buy back stock below NTA ($4.59ps). Whilst a 5% buyback would be ~2.5% accretive to earnings, we assign a low probability to this being completed given current volatility in global return hurdles and general uncertainty on the future outlook for asset values. With competition said to remain very high for direct assets, the group appears to be focussed on development as a way to secure product and supplement growth. Mind the detail in retail holdovers masking headline statistics. Headline retail statistics were generally solid with comparable NPI growth at 3.8% (1H16: 3.0%; FY15: 3.0%) and occupancy at 99.6% (99.4% at June). Whilst the specialty WALE increased to 2.7 years (Dec 15: 2.5 years), specialty expiries in FY17 are now 26% (June: 18%). Whilst GPT indicated holdovers are running at ~6% (static), we note several performance statistics exclude holdovers and given recent administrations, we are remain cautious on returns in the retail sector. Indeed, flat specialty productivity at Highpoint was said to be partially attributable to the Pacific Werribee mall redevelopment (refurbished Myer, new H&M and Uniqlo) albeit we note this is 26km away (likely secondary catchment) and the second international mini-major apparel addition at Charlestown has pulled out (cosmetics replacement). Our work indicates lower international retailer profitability and elevated supply additions. Earnings and target price revision FFO: FY17E: +0.1%; FY18E: +1.1%; FY19E: -0.8% largely reflecting revised interest cost assumptions. NAV +2% to $5.09-$5.27ps. Price catalyst 12-month price target: A$5.09 based on a NAV methodology. Catalyst: GWSCF liquidity review, fund terms to be voted on next week. Action and recommendation Whilst GPT remains our preferred defensive exposure of the predominately retail REITs (over SCG and VCX), with a weaker near-term growth profile and single digit TSR we retain our Neutral recommendation. Please refer to page 20 for important disclosures and analyst certification, or on our website

2 Fig 1 FY16 result directly in line with forecast weaker growth into FY17 also consistent with expectations The good The not-so-good The interesting FY16 result in line with expectations. GPT reported FY16 statutory NPAT of $1,152.7m. FFO of 29.9cps was directly in line with our 29.9cps forecast. The result represents growth of ~5.6% y- y. Drivers in FY16 comprised underlying fixed escalators, lease-up at MLC, a lower debt cost, recognition of the final GWOF performance fee partly offset by dilution from the sale of the Dandenong shopping centre. NTA increased 4.8% or 21cps to $4.59ps as at December 2016 mainly due to favourable property revaluations with a minor tailwind from the negative MTM of interest rate derivatives in 2H16. Maintenance capex and incentives expected to decline. In FY16 maintenance capex and incentives declined -2.6% to $115.5m. This is expected to continue to decline by -4-15% (to $100-$110m) in FY17, driven by lower incentives in the office sub-sector off the back of an improving Sydney office market. We note that this excludes the MLC façade spend. Specialty MAT sales productivity rose ~5.5% to $11,036 per sqm, driving specialty occupancy cost 50bps lower to 16.9%. GPT anticipates that ongoing remixing of the portfolio will lead to increased sales and rents. Occupancy across the retail portfolio rose to 99.6%, up 20bps from 99.4% at June 2016 (99.2% at December 2015). The increase in occupancy was driven by Highpoint (+120bps) and Melbourne Central (+40bps) and Wollongong Central (+100bps). Retail leasing spreads continue to improve, and are now marginally positive (+0.3%), an improvement on the 0% growth in 1H16 and -1.6% in FY15. The group negotiated 504 deals in FY16, with the average annual fixed increase remaining flat at 4.8% (FY15: 4.8%). GPT reported headline net gearing of 23.7% as at 31 December Assuming balance sheet deployment of this capital and taking net gearing to the upper limit of the group s target of 35% allows for ~$2bn of debt-funded acquisition capacity. Assuming a 5.5% initial yield, 96% occupancy and a 3.3% marginal cost of debt, this would result in a ~8% positive impact to group-level earnings. Office NOI of $225.0m is up 6.9%. The benefit of strong LFL growth and the lease up of MLC aided by part period impact of increased stake in GWOF. Comparable income rose 6.3% in FY16 (1H16: 6.0%; FY15: 6.3%; 1H15: 8.1%). Office valuations partially driven by income growth. The group reported total uplift in office valuations of $336.5m or 9.1% which took the total portfolio to ~$4,340.7m (including GPT s stake in GWOF). The positive revaluation movement was 42% attributed by income growth which is a positive outcome. The WACR compressed 39bps to 5.55% over the year. Not acquiring the remaining 50% stake in MLC. GPT mentioned that its capital partner at MLC (QIC) had provided the group with a proposal to acquire the remaining 50% that it did not already own as part of its pre-emptive rights. GPT notified QIC yesterday that it would not proceed with the proposal. We view this as a positive decision by GPT given the lease up of the building to date (99.5% including HoA) is likely to result in limited value creation. Liquidity exists for funds on the secondary market. The group indicated >$850m of secondary units in GWOF and GWSCF (combined) transacted to both new and existing unitholders. Over $500m of units transacted in GWOF. Industrial metrics improve post asset sales and GMF divestment. Like-for-like income growth improved to +1.4% (1H16: +0.1%; FY %) with occupancy increasing to 95.3% (1H16: 92.7%; FY15: 92.3%; June 2015: 92.8%). Headline NOI was ahead of comp NOI growth due to the fullperiod impact of development completions in More components of development pipeline become active. GPT continues to work on its development pipeline which now stands at $3.8bn (up from $3.5bn at 30 June 2016). GPT s share is ~$2bn. Two industrial land sites were acquired during the period with developments now underway. Source: Macquarie Research, February 2017 EPS growth rate into FY17 to slow as expected Consistent with our current forecast for FY17 FFO growth of 1.7%, GPT has guided to 2.0% growth in FY17. The components of our FY17 forecast include the absence of the final GWOF performance fee, Ayers Rock coupon and other transitory items such as the Rouse Hill and Murray Rose sale profits. also gets harder further out due to hedge expiries. Further out (FY20+), we note it gets harder with the base hedge rate increasing to 4.08% vs 3.12% currently (~5.7% all-in using current margins) placing a drag on earnings (MRE FY20: +0.0%). We note the relatively low level of hedging in outer years (4.4 years vs 5.6 years at Dec-15) will eventually place a drag on earnings. Retail NOI of $294.1m was down -0.5% on FY15 (1H16: 0.7%, FY15: 1.2%) primarily due to sale of Dandenong partly offset by increased stake in GWSCF and development income. GPT divested Dandenong Plaza, which was a headwind to FY16 retail earnings (~6% headwind). This was partially offset by an increased stake in GWSCF from 20.2% to 25.3%. The group also received $5.8m in development income (vs $0.8m in FY15) driven by the sale of Rouse Hill development land ($8.1m). FY17 retail lease expiries rise, likely due to short term leases. The next 12 months may prove to be a challenging period for GPT across its retail portfolio with 21% of all leases expiring (includes holdovers) and 26% of specialty leases expiring or expired (includes holdovers). This has increased from June 2016, where total centre FY17 expiries were 14% and specialty expiries were 18%. Comparable specialty sales growth of 2.6% was lower than the 3.7% recorded at September 2016 (4.2% at June 2016; 6.5% at December 2015). Jewellery (+14.5%) and General Retail (+13.9%) were the best performing categories in FY16, with general retail brands driven by cosmetics sales at retailers such as MAC and Mecca. Sales in apparel declined by -1.2% (vs -0.2% at September 2016) as mini-majors sales grew by 12.6%, likely to be due to international mini-majors and JB-HiFi Renewing IAG at CBW was a good outcome but marginally behind expectations. The group extended its existing tenancy (~15,000 sqm) to IAG also has an additional ~13,000 sqm that it can take up at its discretion. The deal was said to have been completed at face rents as per expectations as at acquisition date (September 2014) although the attached incentive was higher than originally expected given greater competing supply - implying net effective rent below expectations. IAG currently occupies 28,520 sqm or ~37% of the office NLA. GWOF acquired 100 Queen St during the period will create more leasing to do. Consistent with media reports over the last few months, GWOF acquired 100% of the current ANZ head office in Melbourne for $274.5m. With the tenant currently occupying 100% of the NLA (34,900sqm), GPT's leasing strategy is to reposition the asset for smaller tenants seeking high quality office space. ANZ will remain there until 2019 before moving to a new building being constructed by LLC in the Docklands. Office occupancy tailwind largely realised. Occupancy in the portfolio declined ~30bps in the last six months to 97.0% (Sep-16: 98.1%; Jun-16: 97.3%; Mar-16: 97.0%; Dec-15: 96.0%) with lease up in 1 Farrer Place and MLC offset by expiries at Melbourne Central Tower and CBW. The actual rent paying occupancy is 93.6% which is below the reported occupancy of 97.0%. The group indicated that actual rent paying occupancy is expected to stay at the 94% level in 2017 given upcoming expiries in 4Q17. Office leasing <1% FFO impact...we have assessed the sensitivity of the group to an improving Sydney office market below. We estimate that there is ~0.3% or ~$1.9m of upside for GPT s earnings (FFO) from positive leasing outcomes in the office portfolio. We have assumed 10% face re-leasing spreads across the expiries in the NSW assets over the next 2 years (FY17-18) and diluted the GWOF impact for GPT s stake (25.3%). We assume the benefit is received in FY17 in its entirety. greater impact to AFFO. While the above analysis is clearly on a face-basis, an improving Sydney CBD office market will experience both face rent increases as well as declining incentives which will see a benefit to AFFO initially. The underlying EPS impact (post amortisation of incentives) will take longer to flow through given typical five-year lease durations and cycling of relatively higher incentives in the last few years. GWSCF fund terms review unitholder vote to occur on 20 February There will be no change to the base fee (45bps of GAV) and the performance fee structure will be removed. The stable base fee is as expected as no performance fee has been payable due to the performance of the fund (this compares to GWOF which saw a 5bps increase in the base fee to 50bps as a partial offset to the removal of the performance fee). GPT will not vote on the new terms with its 25.3% equity stake. The minimum threshold is 75% unitholder votes to approve the new terms. There have been no proxies received yet. Bumping the DPS ahead of the EPS. Interestingly, GPT guided to DPS growth of 5.0% in FY17 (MRE prior: +4.4%). Whilst GPT highlighted its DPS is AFFO covered, the DPS is short of AFFO treating the MLC façade restoration as maintenance activity ($18m spend remaining). Despite this, incentives are expected to fall due to the strong Sydney office market aiding FCF coverage of DPS. Lack of retail stock is likely to lead to an increased focus on developments. As well as limited availability of high quality retail stock, we note GWSCF has first rights on asset acquisitions (and this will not change following the upcoming vote on the funds terms). It is therefore likely that GPT will pursue development opportunities in order to expand its balance sheet retail exposure. M&A is hard to materialise. GPT noted that is continues to review M&A opportunities, however highlighted that it is challenging to achieve the desired outcomes. Consistent with prior research we expect corporate activity to remain elevated reflecting: i) renewed focus on alternate growth levers such as synergy cost out; ii) limited underlying growth; and iii) contracting premiums to NTA across the sector (see: Listed property sector - Walking the tightrope). Buyback will only occur at a discount to NTA... currently trading at a 6% premium. GPT indicated that it will buy back stock if the share price is trading at a discount to NTA, and the group will be no greater than 1/3 of the daily average volume in the stock. Whilst GPT has rolled its buyback, we assign a low probability to the group buying back the full 10% of stock that is entitled to and we do not forecast a buyback in our model. We also note that in order to keep a buyback activated, a group is required notify the ASX approximately every two months. Targeting 3% like for like income growth in retail over the medium term. With specialty tenants currently achieving annual rent increases of ~4.8% on ~56% of the total book this equates to ~2.7% of rental growth in isolation. With income growth across other categories, it is possible to get to a 3.5% rental growth figure. Under an assumption of rising property expenses and outgoings of 2.5%, this leverage results in total NOI growth of ~4.0% (or above the group s target of 3%). Using GPT s definition of AFFO results in AFFOps of 23.5, which is slightly above the DPS of 23.4cps. If we were to re-classify the MLC façade expenditure of FY16 (~$9m) as maintenance capex (considered by GPT as development capex), this would lower AFFO to 22.9cps and would therefore be ~2% below the DPS. Happy with current stakes in wholesale funds. During the period, GPT increased its stakes in GWOF to 24.5% (prior: 20.4%) and GWSCF to 25.3% (prior: 20.2%). While the group did not provide any colour around whether it would increase its holdings in either fund, it did mention that the current stakes are satisfactory although it not rule out participating further in the upcoming GWSCF liquidity review (31 March 2017). The group also indicated that no retail assets are currently considered non-core. Incremental capital deployed in Parramatta. The group acquired a ~2,400 sqm prime site in Parramatta for $31.2m which will see a 26,000 sqm A-grade office tower developed on it for $212m (GPT: 100%). The end value is expected to be >$220m with a yield on cost of >7%. Construction is expected to commence in 2018 with completion in Looking for smaller office tenants. Consistent with recent research (see: Listed property sector - Walking the tightrope) there has been an increase in demand from smaller tenants for office space. This is being welcomed by GPT, which is actively pursuing tenants <10,000 sqm as a form of leasing risk management. When Freehills left MLC occupancy declined to 62%, and now the largest tenant at MLC is just 7% of NLA. 15 February

3 FY17 FFO facing headwinds as a start point Continued headwinds to FY17 earnings growth for GPT comprising: i) absence of the GWOF performance fee; ii) profits from asset sales (Rouse Hill residential site and final release of profits at 3 Murray Rose); and iii) the final coupon from the Ayers Rock Hotel sale (deferred consideration). Whilst these headwinds will be partially offset by a higher base management fee and further trading profits from asset sales (assumed ~$10m pre-tax), we forecast ~2% growth for GPT in FY17. The company has guided to this figure with today s result. Fig 2 FY17 earnings (FFO) bridge FFO bridge - FY16 to FY17e ($m) FY16 FFO Underlying growth Higher Assumed GWOF final GWOF base trading profits perf. fee fees Rouse Hill resi profit 3 Murray Rose sale Final Ayers Rock coupon Other / dev't FY17e FFO Source: Macquarie Research, Company data, February 2017 Strong balance sheet provides a potential tailwind if deployed GPT reported headline net gearing of 23.7% as at 31 December Assuming balance sheet deployment of this capital and taking net gearing to the upper limit of the group s target of 35% allows for ~$2bn of debt-funded acquisition capacity. Fig 3 ~$2bn of deployment capacity with current balance sheet Item Units 31-Dec-16 Deployment Pro-forma Cash $m Intangibles $m Other assets $m 11,726 2,035 13,761 Total assets $m 11,818 13,853 Gross debt $m 2,997 2,035 5,032 Net debt $m 2,940 4,975 Adjustment $m (148) (148) Gearing % 24% 35% Notes: 1. Excludes unrealised fair value adjustments on foreign borrowings. Assuming a 5.5% initial yield, 96% occupancy and a 3.3% marginal cost of debt, this would result in a ~7% positive impact to group-level earnings. There is also a benefit from the wholesale funds in terms of higher distributions from the funds and higher FM and PM fees if the funds both deploy further capital as well (GWOF 17.8% gearing, GWSCF 9.4% gearing). Fig 4 Deployment of balance sheet capacity is ~7% accretive to GPT headstock Capital deployed $m 2,035 Cap rate % 5.50% Occupancy % 96% Passing NPI $m Cost of debt % 3.30% Net interest $m (67.2) FY17 FFO $m Earnings accretion/(dilution) $m 40.3 Earnings accretion/(dilution) % 7.4% 15 February

4 Despite this, GPT indicated pricing for established direct assets was very competitive and a more likely strategy is to increase development activity. Buyback will only occur at a discount to NTA currently trading at a ~6% premium to 31 December 2016 NTA. We understand GPT will buy back stock if the share price is trading at a discount to NTA (not internal NAV). Whilst GPT has recently rolled its buyback, we assign a low probability to the group buying back the full 10% of stock that it is entitled to and we do not model a buyback in our model. The group indicated it will be no greater than 1/3 of the daily average volume in the stock. We also note that in order to keep a buyback activated, a group is required notify the ASX approximately every two months (should it not buyback any stock). We estimate a 5% buyback for GPT to be ~3% accretive to FY17 earnings. This is intuitive given a marginal cost of debt of ~3% and a cost of equity being retired of ~6% (based on the buyback price). There is clearly sufficient balance sheet capacity as per the analysis above. Fig 5 Buying back 10% of the existing share base would result in 6% earnings accretion Shares bought back % 5% 10% GPT buyback price $ps EFPO m 1,798 1,798 Market cap $m 8,253 8,253 Proceeds required $m Marginal cost of debt % 3.2% 3.2% FY17 FFO pre-buyback $m FY17 FFO post-buyback $m EFPO (post-buyback) m 1,708 1,618 FY17 FFOps post-buyback cps FY17 FFOps pre-buyback cps Accretion/(Dilution) % 2.5% 5.6% FFO, AFFO and FCF the impact of capex and incentives With GPT guiding for 5% DPS growth vs 2% FFO growth we have analysed the various measures of dividend sustainability. AFFO ($421.5m) grew by 10.0% in FY16, which was greater than the 7.0% achieved at the FFO level. The divergence of growth between these two metrics was driven by a reduction in maintenance capex and lease incentives by -2.6% (from $118.6m in FY15 to $115.5m in FY16). Using GPT s definition of AFFO results in AFFOps of 23.5, which is marginally above the DPS of 23.4cps. If we were to re-classify the MLC façade expenditure of FY16 of ~$9m as maintenance capex (considered by GPT as development capex), this would lower AFFO to 22.9cps (~2% below the DPS). Fig 6 DPS is covered on a FCF basis, even adjusting for MLC façade spend $m FFO 537 Maintenance capex / incentives AFFO AFFO cps 23.5 MLC façade restoration -9 MRE AFFO MRE AFFO cps 22.9 DPS cps 23.4 FFO 537 Non-cash items 2.7 Other Net operating cashflows Maintenance capex Less incentives Other 13.8 FCF FCF cps 25.2 MLC façade restoration -9 MRE FCF MRE FCF cps 24.7 DPS cps 23.4 Fig 7 but is not covered on an AFFO basis after adjusting for MLC façade spend DPS coverage (%) 10.0% 8% 8.0% 6% 6.0% 4.0% 2.0% 0.2% 0.0% -2.0% -4.0% -2% -6.0% -8.0% -10.0% AFFO MRE AFFO FCF MRE FCF Source: GPT, Macquarie Research, February 2017 Source: GPT, Macquarie Research, February February

5 In a further assessment of DPS coverage, if we were to start with net operating cash flow ($526.2m) and adjust for maintenance capex (-$45.4m), lease incentives (-$41.5m) and other items (+$13.8m), this results in FCF of $453.1m, or 25.2cps, which is ~8% above the DPS. Again, adjusting for the MLC façade spend of $9m, this would suppress FCF to 24.7cps, which is still ~6% greater than the DPS. Whilst we forecast a shortfall next year, the DPS is FCF covered in outer years. We note a material discrepancy between AFFO and FCF in GPT disclosure. Fig 8 DPS well covered by FCF in the half even on an adjusted basis post MLC façade restoration 1H16 2H16 FY16 FY17 FY18 FY19 Cashflow from operating activities EBITDA (incl. associates) Net interest paid (32) (53) (85) (132) (128) (135) Tax paid (6) (6) (6) Other (30) 28 (2) - Net Operating Cashflow Recognition of guaranteed payment for Ayers Rock Maintenance capex and tenant incentives (52) (44) (96) (103) (105) (99) Free Cashflow Distribution to ordinary equity Cash excess / (shortfall) (7) (6) 4 EPS pre distribution to preference shares EPS post distribution to preference shares Free Cashflow per share (cps) Distribution per share EPS yield (post distribution to preference shares) 3.0% 3.0% 6.0% 6.1% 6.4% 6.6% FCF yield 2.4% 2.6% 4.9% 4.8% 5.1% 5.4% Free cash coverage 103.2% 108.2% 105.7% 95.6% 98.7% 100.9% Notes: 1. In operating cash flow we have added back payments for inventory and interest capitalised on developments. 2. We have also excluded rent free incentives as a deduction (in FY16) as operating cash flow is already lower reflecting the absence of rent. 3. MLC facade spend was ~$9m in FY16, we assumed an equal share in the above table. Source: GPT, Macquarie Research, February 2017 Several leasing hurdles to emerge in Melbourne. While the Melbourne office market is clearly improving as indicated in the charts below, GPT has several upcoming expiries/vacancies to deal with. Fig 9 High net absorption in Sydney and Melbourne.. '000 sqm, Net absorption (13) (50) (100) (150) Dec-00 Dec-02 Dec-04 Dec-06 Dec-08 Dec-10 Dec-12 Dec-14 Dec-16 Sydney Melbourne Brisbane Perth Fig has led to net effective rental growth Net effective rent growth (% chg p.a.) 50% 40% 30% 20% 10% -% (10%) (20%) (30%) (40%) (50%) Mar-00 Mar-02 Mar-04 Mar-06 Mar-08 Mar-10 Mar-12 Mar-14 Mar-16 Sydney Melbourne Brisbane Perth 22% 13% (3%) (18%) Source: JLL Research, Macquarie Research, February 2017 Corner of Bourke and William (CBW). GPT and GWOF acquired equal 50% interests in this building in September 2014 from CBUS at a 6.5% cap rate ($608.1m transaction for 100%). The passing yield at the time was said to be ~6% due to a view that the building was under rented. Deloitte represents 24% of the building s area and is leaving to pre-commit MGR s 447 Collins St project. While the expiry is not until May 2020, the loss of the first major expiry in the building was not a good outcome. Including the impact on the co-investment in GWOF as well as GPT s direct investment we anticipate the loss of Deloitte to detract ~1% from GPT s earnings on a full-year basis. 15 February

6 although IAG was a solid outcome with the group extending its existing tenancy announced with today s result (~15,000 sqm) to IAG also has an additional ~13,000 sqm that it can take up at its discretion. We highlight IAG currently occupies 28,520 sqm or ~37% of the total office NLA which means IAG is only committing to ~50% of its current space. The deal was said to have been completed at face rents as per expectations as at acquisition date (September 2014) although the attached incentive was higher than originally expected given greater competing supply implying net effective rent below asset acquisition feasibility expectations. The building s book value increased by 4.8% or $15.3m (GPT 50% share) during the six months to 31 December with the IAG deal said to be the main driver of this movement (~4% or ~$13m) which conflicts with the leasing outcome. AMP at 750 Collins St. GWOF has a 100% interest in 750 Collins St Melbourne. The building is 100% leased to AMP, with the lease expiring in November Again, while a longerdated risk (i.e. FY20), allowing for GPT s stake in GWOF the impact of AMP leaving would again be ~1% of earnings from this lease in isolation. GWOF acquired 100 Queen St during the period will create more leasing to do. Consistent with media reports over the last few months, GWOF acquired 100% of the current ANZ head office in Melbourne for $274.5m. With the tenant currently occupying 100% of the NLA (34,900sqm), GPT s leasing strategy is to reposition the asset for smaller tenants seeking high quality office space. ANZ will remain there until 2019 before moving to a new building being constructed by LLC in the Docklands. While office conditions are currently improving in Melbourne, we note GPT will have to lease ~120,000 sqm of space in the Melbourne market in 2020/2021. This represents ~2.5% of the market and will occur at a time of increasing supply additions as indicated in the chart below. Fig 11 Melbourne office stock is expected to grow by ~2.5% between CY17-CY18 Stock additions / total stock - Melbourne (%) 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% (1.0%) (2.0%) (3.0%) Average = 2.2% (4.0%) Dec-87 Dec-92 Dec-97 Dec-02 Dec-07 Dec-12 Dec-17 JLL F'cast 1.9% 0.6% Melbourne 20-Yr Average Source: JLL, Macquarie Research, February 2017 GPT remains exposed to an improving Sydney office leasing environment In our recent research (see: Listed property sector - Walking the tightrope, page 96), we highlighted the strength of the Sydney CBD office market compared to the other major CBD office markets and an expectation for this to continue given limited new supply in Sydney until We have assessed the sensitivity of the group to an improving Sydney office market below. We estimate that there is ~0.3% or ~$1.9m of upside for GPT s earnings (FFO) from positive leasing outcomes in the office portfolio. We have assumed 10% face re-leasing spreads across the expiries in the NSW assets over the next two years (FY17-18) and included the GWOF impact for GPT s stake (25.3%). We assume the benefit is received in FY17 in its entirety. 15 February

7 Fig 12 10% re-leasing spreads in NSW over the next 2 years has a ~0.3% impact to FFO Units GPT GWOF Total NPI $m Lease expiries (FY17-18) % 17% 6% NPI exposed $m Spread % 10% 10% NPI benefit $m Exposure % 100.0% 25.3% NPI benefit (ownership-adjusted) $m FY17 FFO $m Accretion/(Dilution) % 0.3% We have shown the sensitivity of group-level earnings to a change in the re-leasing spread in the table below given uncertainty around the re-leasing spreads that will be achieved. Fig 13 Sensitivity of group earnings to re-leasing spreads in NSW Face net re-leasing spreads (5.00%) 0.00% 5.00% 10.00% 15.00% 20.00% 25.00% (0.2%) -% 0.2% 0.3% 0.5% 0.7% 0.8% While this is clearly on a face-basis, an improving Sydney CBD office market will experience both face rent increases as well as declining incentives which will see a benefit to AFFO initially. The underlying EPS impact (post amortisation of incentives) will take longer to flow through given typical five-year lease durations and cycling of relatively higher incentives in the last few years. We have disaggregated the drivers of effective rental growth in Sydney CBD over the last ~24 months which indicate the bulk of the growth is due to face rent increase and a partial contribution from declining incentives (refer chart below). Fig 14 Contributions to net effective rental growth in Sydney CBD Contributions to net effective rent growth (%) 22.5% 25.0% 18.4% 20.0% 15.8% 17.3% 15.0% 12.7% 7.7% 14.5% 17.9% 10.0% 10.1% 10.8% 12.3% 13.0% 1.6% 0.6% 5.0% 3.9% 5.5% 2.2% 3.4% 4.0% 5.2% 6.0% -% (2.6%) (3.1%) (2.0%) (5.0%) (10.0%) Mar-15 Jun-15 Sep-15 Dec-15 Mar-16 Jun-16 Sep-16 Dec-16 Incentives Rental growth Outgoings Source: JLL Research, Macquarie Research, February 2017 Recap on potential implications from residential site sales In February 2016, GPT identified several assets that may be eligible for urban renewal opportunities including higher and better use (residential development). These assets include: i) Rosehill Business Park, Camellia; ii) Town Centre, Sydney Olympic Park; and iii) Rouse Hill Town Centre. We had reviewed the assets and the financial impact in greater detail in our original research here: GPT Group - Block party, April February

8 Fig 15 Property metrics for assets identified for change in use opportunities Rosehill Business Park, Camellia Town Centre, Sydney Olympic Park Rouse Hill Town Centre Metric Units Valuation (100%) $m Asset class Industrial Industrial Retail GPT share % Cap rate % Site area sqm 79,700 52,900 n.p. GLA sqm 41,900 26,500 69,300 Identified apartments # 3,000 1,991 1,000 WALE Years n/a Notes: 1. Metrics as at 31 December The number of identified apartments was as per GPT commentary at its February 2016 final results conference call with the ratio for SOP extrapolated based off the GLA/apartment ratio at Camellia. Source: GPT, Macquarie Research, February 2017 The update as at 31 December 2016 includes: i) a draft rezoning to be exhibited in mid-2017 for Rosehill (Camilla) and finalised in 1H18; and ii) a masterplan gazettal expected in 2H17 with authorities (Sydney Olympic Park Authority (SOPA)) still finalising wider infrastructure and transport plans. On balance, we remain of the view that the group will realise value through a sale of the land/assets to a developer (consistent with our original research here: - Block party, April 2016). The rationale behind these structures is a desire to be prudent on risk. With recent hires in the business from competing residential development firms, GPT is building internal capacity in this regard. Fig 16 Summary of apartment outcomes...~27-32cps depending on preferred structure Strategy Units Rosehill Business Park, Camellia Town Centre, Sydney Olympic Park Rouse Hill Town Centre Self-develop cps Sale to developer cps Notes: 1. Represents present value of apartment profits. FY16 result in line with expectations GPT reported FY16 statutory NPAT of $1,152.7m. FFO of 29.9cps was directly in line with our 29.9cps forecast. The result represents growth of ~5.6% y-y. Drivers in FY16 comprised underlying fixed escalators, lease-up at MLC, a lower debt cost, recognition of the final GWOF performance fee partly offset by dilution from the sale of the Dandenong shopping centre. Whilst the change in segment reporting is again complicating comparison to divisional forecasts, net income across the divisions of $675.5m (includes co-investment income and precorporate/interest) was in line with our expectations, with net interest lower than expectations and income tax slightly higher (GWOF fees etc. were taxable in FM). Total 15 February

9 Fig 17 FY16 FFOps increased by 5.6% on the pcp driven by lease up of MLC, asset recycling profits booked in the retail and industrial divisions and GWOF performance fee Jun-15 Dec-15 Dec-15 Jun-16 Dec-16 Dec-16 Item 1H15 2H15 FY15 1H16 2H16 FY16 Change Retail $m (0.5%) Office $m % Logistics $m % Funds Management $m % Net Income $m % Net finance costs $m (57.3) (58.6) (115.9) (50.1) (49.9) (100.0) (13.7%) Corporate overheads $m (16.4) (16.7) (33.1) (13.8) (16.0) (29.8) (10.0%) Tax expenses $m (5.7) 0.8 (4.9) (5.9) (8.1) (14.0) 185.7% Non-core $m (49.0%) Funds from Operations (FFO) $m % Property revaluations $m % MTM of interest rate derivatives $m 7.3 (81.3) (74.0) (65.7) 42.7 (23.0) (68.9%) Other $m 19.6 (11.3) % Statutory NPAT $m , % FFO cps % DPS cps % Notes: 1. Net income items include development income. Retail (48% of asset mix) Solid operating metrics but FY17 lease expiries provides a risk Retail NOI of $294.1m was down -0.5% on FY15 (1H16: 0.7%, FY15: 1.2%) primarily due to the sale of Dandenong partly offset by increased stake in GWSCF and development income. GPT divested Dandenong Plaza, which was a headwind to FY16 retail earnings (~$17m, or 6% of earnings). This was partially offset by an increased stake in GWSCF from 20.2% to 25.3% in 2H16. The group also received $5.8m in development income (vs $0.8m in FY15) driven by the sale of Rouse Hill residential development land ($8.1m). Fig 18 Specialty sales growth moderated by 390bps across FY16 to 2.6% Retail 1H15 FY15 1H16 FY16 Change on pcp NOI A$m (2.0%) Comparable income growth % 3.2% 3.0% 3.0% 3.8% 80bps Occupancy % 99.4% 99.2% 99.4% 99.6% 40bps Comparable total centre sales growth % 3.7% 4.1% 3.2% 3.2% (90bps) Comparable specialty sales growth % 5.9% 6.5% 4.2% 2.6% (390bps) Specialty occupancy costs % 17.8% 17.4% 17.1% 16.9% (50bps) Average re-leasing spread % (3.1%) (1.6%) -% 0.3% 190bps Notes: 1. Part of the reason for the specialty sales slowdown (MAT for comparable malls) reflected a space down weight towards more mini-major categories. Source: GPT, Macquarie Research, February 2017 Targeting 3% like for like income growth in the medium term. GPT outlined a medium term like for like income growth target of 3%. With specialty tenants currently achieving annual rent increases of ~4.7% (on ~56% of the book assuming specialty accounts for 70% of income and ~20% expires per annum), this equates to ~2.7% of rental growth in isolation. Adding in leasing spreads of +0.6% across ~20% expiring specialty book, as well as rental growth across anchor tenants, mini-majors and ancillary income, it is possible to get to a 3.5% rental growth figure. Under an assumption of rising property expenses and outgoings of only 2.5%, this leverage results in total NOI growth of ~4.0% (which is above the group s target of ~3%). 15 February

10 Fig 19 GPT is targeting 3% LFL retail income growth possible with 3.5% rent growth and 2.5% cost growth Growth Growth rate assumption Proportion of income assumption Comment Specialty escalators 2.7% 4.8% 56% Assumes spec is 70% of income and 20% Specialty leasing spreads 0.1% 0.6% 14% lease expiry profile Anchor tenants 0.2% *1.0% *15% *MRE assumption Mini-majors 0.2% 4.3% *5% *MRE assumption Supplementary income 0.4% 4.0% 10% Total 3.5% 100% Rent (FY16) 349 Rental growth (at 3.5%) 12 Total Rent 361 Property expenses & outgoings (FY16) (102) Property expenses & outgoings growth (at 2.5%) (3) Total property expenses & outgoings (105) Total NOI (with growth) 256 FY16 NOI 247 NOI growth 4.0% $m Note: 1. MRE assumptions for proportion of anchor tenant and mini-major income. 2. MRE assumptions for anchor tenant rental growth rate. 3. Assuming specialty accounts for 70% of income and has an expiry profile of 20% per annum. Source: GPT, Macquarie Research, February 2017 The above analysis assumes stable occupancy as well as no down-time between lease expiry and lease renewal. Although this analysis does have its limitations, it provides an indication of how GPT may be able to reach its internal 3% growth target. Occupancy across the portfolio rose to 99.6%, up 20bps from 99.4% at June 2016 (99.2% at December 2015). The increase in occupancy was partially driven by a reduction in vacancy at Highpoint (+120bps), Melbourne Central (+40bps) and Wollongong Central (+100bps). The group has been able to backfill the space voided by Myer at Wollongong Central through the addition of David Jones (not yet opened) as well as the opening of H&M, Rebel and Anaconda in 2H16. The above mentioned increase in occupancy has been offset by declining occupancy at Chirnside (- 110bps), Norton Plaza (-90bps) and Parkmore Shopping Centre (-80bps). Specialty MAT sales productivity rose ~5.5% to $11,036 per sqm, driving specialty occupancy cost 50bps lower to 16.9%. GPT anticipates that on-going remixing of the portfolio will lead to increased sales and rents, however we have previously written on the headwinds facing the retail sector and in-turn retail landlords (see: Listed property sector - Walking the tightrope). Leasing spreads continue to improve, and are now marginally positive (+0.3%), an improvement on the 0% growth in 1H16 and -1.6% in FY15. The group negotiated 504 deals in FY16, with the average annual fixed increase remaining flat at 4.8% (FY15: 4.8%). Comparable specialty sales growth of 2.6% was lower than the 3.7% recorded at September 2016 (4.2% at June 2016; 6.5% at December 2015). Jewellery (+14.5%) and General Retail (+13.9%) were the best performing categories in FY16, with general retail driven by cosmetics sales at retailers such as MAC and Mecca. Sales in apparel declined by -1.2% (vs -0.2% at September 2016) which is was driven by: i) increased remixing away from apparel; and ii) increased competition from mini-majors. The group did note that on a per sqm basis, apparel sales have increased by 5.2%. The mini major category achieved growth of 12.6%, which is an increase on the 8.9% achieved at September 2016 (likely due to internationals and better performing tenants such as JB Hi-Fi). Although the MAT growth of the apparel mini-majors are initially strong, we continue to highlight sales per store decline as the asset stabilises (see: Listed Property Sector - Uniqlo needs some glow). We continue to highlight the headwinds created by the international mini-majors on domestic apparel retailers and in turn retail landlords (see: Listed property sector - Walking the tightrope). Sales growth for the department stores has remained strong, with sales up 4.2% vs 4.3% in September 2016 (June 2016: 3.7%; December 2015: 1.9%) while DDS sales growth remains anaemic (Dec-16: -2.6%; Sep-16: -0.7%; June-16: +0.7%; March-16: -0.7%). FY17 lease expiries rise due to holdovers. The next 12 months may prove to be a challenging period for GPT across its retail portfolio with 21% of all leases expiring (includes holdovers) and 26% of specialty leases expiring or expired (includes holdovers). 15 February

11 FY17 total centre and specialty lease expiries have increased over the past six months (see below charts), driven by rising holdovers. For example, as at 1H16 specialty lease expiries in FY16 were sitting at 16% and 2017 lease expiries were 18%. With FY17 specialty lease expiries rising to 26%, this implies that holdovers were responsible for the increase in FY17 lease expiries as at FY16. Whilst GPT indicated some of these holdovers are enforced by GPT as they seek a better tenant or commence development etc, we believe it is a sign that tenant and landlord can t agree terms on new leases. Fig 20 21% of total centre leases expire in FY17 Lease expiries - Total centres (%) 40% 35% 30% 25% 20% 15% 10% 5% -% 12% 21% 14% 15% 16% 13% 13% 15% 15% 35% 31% H16 FY16 Fig 21 26% of specialty leases expire in FY17 Lease expiries - Specialties (%) 30% 26% 27% 25% 20% 18% 16% 17% 17% 18% 16% 15% 16% 15% 15% 10% 5% -% H16 FY16 Source: GPT, Macquarie Research, February 2017 Source: GPT, Macquarie Research, February 2017 Several of the assets that have a high level of leases expiring also have very high levels of sales productivity (eg Charlestown Square, Sunshine Plaza and Westfield Penrith) although there are also a couple that will likely prove problematic given much lower sales productivity (eg Northland and Parkmore), particularly given the specialty occupancy cost ratio is ~19%. Fig 22 Northland and Parkmore screen as having heightened leasing risk Fig 23 Specialty WALE has remained stable at 2.7 years Specialty productivty ($/sqm) FY17 lease expiry (%) 14,000 12,000 10,000 8,000 6,000 4,000 2,000-40% 35% 30% 25% 20% 15% 10% 5% 0% Specialty WALE (Years) Specialty MAT ($/sqm) FY17 lease expiry - Dec-11 Jun-12 Dec-12 Jun-13 Dec-13 Jun-14 Dec-14 Jun-15 Dec-15 Jun-16 Dec-16 Source: GPT, Macquarie Research, February 2017 Source: GPT, Macquarie Research, February 2017 The stabilisation in the specialty WALE at 2.7 years is a positive, in light of the increase in expiries in FY17. On the result call management indicated that headwinds still remain to the retail environment and noted that 16 retailers (64 tenancies) folded during in FY16. Although the group was able to increase occupancy despite these closures, we believe the structural issues facing the retail environment (see: Listed property sector - Walking the tightrope) will continue to have an impact on specialty leasing spreads, especially in light of the groups elevated FY17 lease expiry profile. The retention rate rose to 75%, compared to 69% at 1H16 (and 70% at FY15). This heightened retention rate is likely to be influenced by holdovers. 15 February

12 Total retail developments (underway, planned and future) are ~$1.2bn at December Along with developments underway (Macarthur Square shopping centre estimated total development cost $240m, GWSCF 50% share) and Sunshine Plaza ($400m, GPT 50%), GPT has a number of developments due to commence construction shortly. This includes a $250m expansion of Rouse Hill Town Centre (to commence in 1H17, GPT 100%) and a $110m development at Melbourne Central (GPT: 100%). The future development pipeline also includes a $334m development at Highpoint. Rouse Hill development revised. The development cost of Rouse Hill has declined by ~$50m as a result of: i) a change in scope of the development; and ii) the exclusion of the residential portion of the development from these costs (GPT expects to incorporate ~70 apartments into the development). GPT still aiming for a stabilised yield on cost of >6% at Sunshine Plaza. Consistent with recent research (see: Listed property sector - Walking the tightrope), we remain cautious of development returns, especially on assets that are delivering a yield on cost similar to its current cap rate. Indeed, the expected yield on cost at Sunshine Plaza is just 50bps above the current cap rate (the cap rate declined by 25bps to 5.50% in the six months to December 2016). Positively, the development at Sunshine Plaza is a bolt-on and therefore will have limited impact to trading at the mall during development (ie. limited downtime). Office (39% of asset mix) Occupancy build supports growth Office NOI of $225.0m is up 6.9% against FY15 (1H16: +0.0%; FY15: 8.5%; 1H15: 8.8%) with the benefit of strong LFL growth and the lease up of MLC aided by part period impact of increased stake in GWOF. Comparable income rose 6.3% in FY16 (1H16: 6.0%; FY15: 6.3%; 1H15: 8.1%). Fig 24 Office NOI growth driven by strong LFL growth on the back of lease up Office FY15 1H16 FY16 Change on pcp NOI A$m % Comparable income growth % 6.3% 6.0% 6.3% 0bps Occupancy % 96.8% 97.3% 97.0% 20bps WALE years (0.7) Occupancy in the portfolio declined ~30bps in the last six months to 97.0% (Sep-16: 98.1%; Jun-16: 97.3%; Mar-16: 97.0%; Dec-15: 96.0%) with lease up in 1 Farrer Place (FY16: 91.7%; 1H16: 77.8%) and MLC (FY16: 99.5%; 1H16: 97.8%) offset by expiries at Melbourne Central Tower (FY16: 94.9% leased; 1H16: 99.3% leased) and CBW (FY16: However we note that these reported occupancy numbers include signed leases and are not reflective of in-place income. The actual rent paying occupancy is 93.6% which is below the reported occupancy of 97.0% (including 3.4% of signed leases) and heads of agreement-occupancy of 98.1%. The group indicated that actual rent paying occupancy is expected to stay at the 94% level given upcoming expiries in 4Q17. From an income security perspective, only ~15% of GPT s office leases expire in the next 24 months which we view as a positive. Although, we highlight lease expiries in 2020 and 2021 are elevated which occur in a period that is matched against expected supply completions. 15 February

13 Fig 25 Only 15% of office leases expire in next 2 yrs Fig 26 Occupancy build due to 1 Farrer Place Lease expiries - Office (%) 20% 19% 18% 17% 16% 14% 14% 12% 11% 10% 9% 8% 8% 7% 6% 6% 6% 4% 3% 2% -% Occupancy change over six months (%) 15.0% 10.0% 5.0% -% (5.0%) (10.0%) 0.6% 0.1% Australia Square Citigroup Centre 1.7% MLC 13.9% 1 Farrer Place (4.4%) Melbourne Central Tower (2.0%) CBW 0.2% One One One Eagle St Notes: 1. Includes signed leases Notes: 1. Includes signed leases The key expiries over the next couple of years are provided below with Melbourne Central Tower, Australia Square and Riverside Centre (QLD) the biggest risks to GPT headstock. Fig 27 Assets with >25% expiries in the next 2 years GPT Asset Expiry next 24 months (%) Estimated income (per annum $m) Income at risk ($m) Australia Square 27% Melbourne Central Tower 27% GWOF HSBC Centre 31% Southbank Boulevard 46% Collins St 27% Riverside Centre 35% Queen St 79% Total 41.0 Not acquiring the remaining 50% stake in MLC. GPT mentioned that its capital partner at MLC (QIC) had provided the group with a proposal to acquire the remaining 50% that it did not already own as part of its pre-emptive rights. GPT notified QIC yesterday that it would not proceed with the proposal. We view this as a positive decision by GPT given the lease up of the building to date (99.5% including HoA) is likely to result limited value creation. Valuations partially driven by income growth. The group reported total uplift in office valuations of $336.5m or 9.1% which took the total portfolio to ~$4,340.7m (including GPT s stake in GWOF). The valuations were 42% driven by income growth which is a positive outcome. The WACR compressed 39bps to 5.55% over the year. Consistent with our recent research (see: Listed property sector - Walking the tightrope, page 96), GPT referenced that the Sydney and Melbourne office markets are showing stronger signs of tenant demand. This strong demand combined with limited net supply over the next 3 years will lower vacancy and support continued growth in effective rents. We believe GPT should be able to re-lease space at improved spreads, especially moving into 2017 and 2018 (refer to our re-leasing spread analysis above). The group referenced its technology tenant base are looking to grow space. 15 February

14 Fig 28 Entering an era of limited supply %, Sydney CBD vacancy (by area) 12.0% 9.8% 10.0% 7.7% 8.0% 7.5% 6.0% 5.4% 4.8% 5.1% 3.9% 4.0% 3.1% 3.3% 2.9% 5.1% 2.0% -% 1.9% 1.8% 1.1% End 2016 End 2017 End 2018 End 2019 End 2020 End 2021 Low Mid High Fig 29 Sydney leasing demand strong '000 sqm, Net absorption (Sydney) (50) (100) (150) Dec-97 Dec-00 Dec-03 Dec-06 Dec-09 Dec-12 Dec-15 Annual net absorption 20-Yr Average Source: JLL Research, Macquarie Research, February 2017 Source: JLL Research, Macquarie Research, February 2017 Industrial (13% of asset mix) LFL income growth improving Industrial operating income of $95.4m is up 0.8% on FY15 (1H16: -3.3% FY15: 6.4%). Like-for-like income growth improved to +1.4% (1H16: +0.1%; FY %) with occupancy increasing to 95.3% (1H16: 92.7%; FY15: 92.3%; June 2015: 92.8%). Headline NOI was ahead of comp NOI growth due to the full-period impact of development completions in Fig 30 Industrial soft but better than prior periods metrics better due to asset sales Industrial FY15 1H16 FY16 Change on pcp NOI A$m % Comparable income growth % 0.7% 0.1% 1.4% 70bps Occupancy % 92.3% 92.7% 95.3% 300bps WALE years (0.3) Headline NOI will also be supplemented in 2017 by the expected completion of four industrial assets with a total forecast cost of $125m. Three of these assets are expected to complete in 1H17 with one in 2H17. Refer table further below for greater detail on the development pipeline. There has been solid progress on reducing the expiry profile with ~24% of expiries in FY17 and FY18, down from 27% as at June Fig 31 ~24% of industrial expiries in the next 24 months Lease expiries - Industrial (%) 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% -% 17% 13% 11% 10% 11% 12% 12% 9% Source: GPT, Macquarie Research, February % 3% 3% 2% 3% 3% 44.0% 40.0% H16 FY16 Funds management limited FUM growth in future periods GPT reported $61.0m of funds management net income (including the fund distributions), up 36.8% on the pcp on the back of 4% growth in FUM (to $10.4b) and GWOF performance fees. Fee income was also supported by a higher base management fee from the GWOF vehicle (effective 1 July 2016). 15 February

15 FUM growth was predominantly in 1H16 due to strong property revaluations and developments while 2H16 saw headwinds to FUM growth due to asset divestments (Westfield Woden). We forecast FM income to decline in FY17 as the group cycles the GWOF performance fee. Fig 32 Funds management income supported by GWOF performance fees GWOF GWSCF Unit Dec-15 Jun-16 Dec-16 Dec-15 Jun-16 Dec-16 Number of assets # Property value A$bn Weighted average cap rate % 6.03% 5.68% 5.55% 5.73% 5.73% 5.46% Gearing % 14.8% 13.7% 17.8% 14.2% 14.2% 11.5% GPT ownership interest % 20.4% 20.4% 24.5% 20.2% 20.2% 25.3% A$m 980 1,042 1, month total return (post-fees) % 14.9% 18.6% 14.5% 4.4% 4.4% 11.5% Source: GPT, Macquarie Research, February 2017 Increased stakes in both funds. During the period, GPT increased its stakes in GWOF to 24.5% (prior: 20.4%) and GWSCF to 25.3% (prior: 20.2%). Our previous research had indicated that both stakes would be ~1% accretive to GPT s earnings (higher co-investment income). While the group did not provide any colour around whether it would increase its holdings in either fund, it did mention that the current stakes are satisfactory although it not rule out participating further in the upcoming GWSCF liquidity review (31 March 2017). GWSCF fund terms review unitholder vote to occur on 20 February The terms of GWSCF were said to be well progressed with no change to the base fee (45bps of GAV) although the performance fee structure will be removed (despite no performance fee payable due to the performance of the fund). All other key terms are expected to remain unchanged. GPT will not vote on the new terms with its 25.3% equity stake. The minimum threshold is 75% unitholder votes to approve the new terms. There have been no proxies received yet. GWSCF liquidity review window opening in March. This is a 10-year liquidity review window and we note the fund has been a poor performer including a problematic retail development project at Wollongong Central (20% writedown). GPT invested $157m in GWSCF last year and there is a risk of elevated redemptions in March that may require a further stake increase from GPT, gearing increase, asset sales etc to fund any shortfall. Liquidity on the secondary market. The group indicated >$850m of secondary units in GWOF and GWSCF (combined) transacted to both new and existing unitholders. Over $500m of units transacted in GWOF which implies $350 in GWSCF. No longer managing GMF. During 1H16, GPT sold its 13% interest in GMF to GOZ. The group has also entered into a facilitation deed ($9m) for the transfer of the management rights to GOZ. GPT received $2.0m of management fees and $1.4m of FFO for managing GMF in FY16. GPT ceased to manage GMF from 30 September Development pipeline increases to $3.8bn GPT continues to work on its development pipeline which now stands at $3.8bn (up from $3.5bn at 30 June 2016). GPT s share is ~$2bn. Main changes to the pipeline include: Splitting out the residential component of the Rouse Hill development which explains the $50m decline. The residential component ($50m) is not included in the pipeline but is expected to yield ~70 apartments. Acquisition of a ~2,400 sqm prime site in Parramatta for $31.2m which will see a 26,000 sqm A-grade office tower developed on it for $212m (GPT: 100%). The end value is expected to be >$220m with a yield on cost of >7%. Construction is expected to commence in 2018 with completion in February

16 Dec-16 Jun-17 Dec-17 Jun-18 Dec-18 Jun-19 Dec-19 Jun-20 Dec-20 Jun-21 Dec-21 Macquarie Wealth Management Fig 33 $3.8bn development pipeline acquired Parramatta site for office development Sub-sector FY15 1H16 FY16 FY16 vs 1H16 Underway Macarthur Square Retail % Sunshine Plaza Retail % Wollongong Central Retail (3%) 4 Murray Rose Avenue Office % Lot 2012 Eastern Creek Drive, Eastern Creek Industrial % 1A Huntingwood, Huntingwood Industrial Whitelaw Place, Richlands Industrial Abbott Road, Seven Hills Industrial % Total Underway % Planned Rouse Hill Retail (17%) Melbourne Central Retail % Chirnside Retail (15%) Casuarina Square Retail % MLC Office (53%) Phillip Street & 32 Smith Street, Parramatta Office 212 Austrak Minto Industrial % Lot 21, Old Wallgrove Road, Eastern Creek Industrial % Erskine Park (1 Lockwood Rd, Templar Rd) Industrial (70%) Erskine Park (Lot 11, Templar Rd) Industrial % Austrak Somerton Industrial % 1B Huntingwood, Huntingwood Industrial 19 Metroplex, Wacol Industrial (9%) Wembley Industrial (9%) Total planned 1,054 1,094 1,221 12% Future pipeline Highpoint Retail % Parkmore Retail % Other 1,650 1,530 1,620 6% Total future pipeline 2,014 1,894 1,984 5% Total development 3,474 3,539 3,818 8% Successful delivery of these developments would be expected to be earnings-accretive for GPT in light of the low interest rate environment which currently prevails. Notably, with a relatively narrow spread between development initial yields and cap rates of existing assets, we expect total development returns will be increasingly underpinning by final cap rate compression. GPT s average cost of debt in FY16 was 4.25%, which is ~5bps lower than at June 2016 and ~35bps lower than December 2015 (4.60%). This was due to lower fixed and lower floating interest rates and longer duration debt (higher margin). Hedging is expected to average 53% in This is expected to increase to 66% in 2018 with the group expecting a small risk of interest rate increases over the medium term. The group expects the WACD to remain steady at 4.25% in 2017 with higher reference rates offset by more favourable bank margins. Fig 34 Conservative balance sheet metrics Fig 35 Hedge base rates increase in Dec Jun Dec-16 Total debt 2,948 2,792 2,997 Gearing 26.3% 24.4% 23.7% Look-through gearing 27.8% 26.2% 25.7% Interest cover ratio 5.3x 6.3x 6.4x Weighted average cost of debt 4.60% 4.30% 4.25% Weighted average debt maturity Credit rating (S&P/Moody's) A-/A3 A/A3 A/A3 1. Net debt total tangible assets less cash 2. Look-through net debt look-through total tangible assets less cash 3. EBIT gross finance costs excluding capitalised interest 4. Includes fees and margins 5. S&P/Moody s Average rate on hedged balance ex-margins (%) 4.50% 4.00% 3.50% 3.00% 2.50% 2.00% Medium term hedge rates have increased dramatically... As at 30-Jun-16 As at 31-Dec-16 NTA per security increased 4.8% in the last six months NTA increased 4.8% or 21cps to $4.59ps as at December 2016 mainly due to favourable property revaluations with a minor tailwind from the negative MTM of interest rate derivatives in the half. 15 February

17 Fig 36 Cap rates fell by 31bps across GPT s portfolio vs the pcp Weighted average cap rate (%) Fair value ($m) Average cap rate change Average value change Dec-15 Jun-16 Dec-16 Dec-15 Jun-16 Dec-16 1H17 vs 1H16 1H17 vs 2H16 1H16 vs 1H15 1H16 vs 2H15 Retail 5.58% 5.52% 5.39% 5,044 4,939 5,318 (19bps) (13bps) 5% 8% Office 5.94% 5.58% 5.55% 3,710 4,034 4,341 (39bps) (3bps) 17% 8% Industrial 7.03% 6.81% 6.54% 1,349 1,437 1,405 (49bps) (27bps) 4% (2%) Total/Weighted average 5.91% 5.72% 5.60% 10,103 10,411 11,063 (31bps) (12bps) 10% 6% Source: GPT, Macquarie Research, February February

18 Fig 37 GPT financial summary Item (A$m) FY15 FY16 FY17 FY18 FY19 FY20 Revenue Rent from investment properties Revenue from hotel operations Property and fund management fees Development project revenue Development profits Total segment revenue Other Income Share of after tax profits of investments in associates and JVs Interest revenue - associates and other investments Total other income Total segment revenue and other income Implied operating costs (excl. mgmt and other admin costs) (161.7) (170.8) (158.1) (161.9) (166.0) (169.5) EBITDA (excl. mgmt and other admin costs) Management and other administration costs (73.5) (68.0) (62.9) (64.3) (65.9) (67.3) Depreciation and amortisation expense Finance costs (117.7) (102.6) (121.9) (118.0) (125.4) (147.6) Income tax (expense) / benefits (5.5) (14.0) (6.2) (6.0) (6.4) (7.5) Other Segment result for the year Other adjustments to reach net gain/loss for the year Distribution to exchangeable securityholders (1.7) FFO Adjusted NPAT FFO per share FFO growth 5.5% 5.7% 1.9% 5.2% 2.8% 0.2% EPS EPS growth 5.6% 2.6% 2.1% 5.7% 3.1% 0.2% DPS DPS growth 6.1% 4.0% 5.1% 5.5% 2.8% 0.2% Balance sheet summary FY15 FY16 FY17 FY18 FY19 FY20 Current Assets Cash & cash equivalents Non-current assets classified as held for sale Other current assets Total current assets Non-current assets Investment properties 7, , , , , ,353.7 Investment in JV and associates 2, , , , , ,120.2 Other non-current assets Total non-current assets 10, , , , , ,085.9 Total assets 11, , , , , ,330.4 Liabilities Borrowings 2, , , , , ,252.8 Other liabilities Total liabilities 3, , , , , ,610.2 Net assets 7, , , , , ,720.2 Source: Macquarie Research, Company Data, February 2017 Stocks mentioned Scentre Group (SCG AU, A$4.43, Underperform, TP: A$4.51) Vicinity Centres (VCX AU, A$2.89, Underperform, TP: A$2.93) 15 February

19 Macquarie Quant View The quant model currently holds a neutral view on. The strongest style exposure is Valuations, indicating this stock is under-priced in the market relative to its peers. The weakest style exposure is Price Momentum, indicating this stock has had weak medium to long term returns which often persist into the future. 539/1032 Global rank in Real Estate % of BUY recommendations 40% (4/10) Number of Price Target downgrades 1 Number of Price Target upgrades 1 Fundamentals Attractive Quant Local market rank Global sector rank Displays where the company s ranked based on the fundamental consensus Price Target and Macquarie s Quantitative Alpha model. Two rankings: Local market (Australia & NZ) and Global sector (Real Estate) Macquarie Alpha Model ranking A list of comparable companies and their Macquarie Alpha model score (higher is better). Factors driving the Alpha Model For the comparable firms this chart shows the key underlying styles and their contribution to the current overall Alpha score. Scentre Group 0.2 Scentre Group Dexus Property Group 0.1 Dexus Property Group Investa Office Fund 0.1 Investa Office Fund % -80% -60% -40% -20% 0% 20% 40% 60% 80% 100% Valuations Growth Profitability Earnings Momentum Price Momentum Quality Macquarie Earnings Sentiment Indicator The Macquarie Sentiment Indicator is an enhanced earnings revisions signal that favours analysts who have more timely and higher conviction revisions. Current score shown below. Drivers of Stock Return Breakdown of 1 year total return (local currency) into returns from dividends, changes in forward earnings estimates and the resulting change in earnings multiple. Scentre Group Dexus Property Group Scentre Group Dexus Property Group Investa Office Fund -0.4 Investa Office Fund % -30% -20% -10% 0% 10% 20% 30% 40% Dividend Return Multiple Return Earnings Outlook 1Yr Total Return What drove this Company in the last 5 years Which factor score has had the greatest correlation with the company s returns over the last 5 years. PEG Ratio Inverted Price to Cash LTM Price to Earnings NTM Price Upside Net Income Margin NTM Change in PPE FY0 Momentum 6 Month Momentum 12 Month Negatives Positives -25% -14% -15% -18% 29% 29% 28% 36% -40% -20% 0% 20% 40% How it looks on the Alpha model A more granular view of the underlying style scores that drive the alpha (higher is better) and the percentile rank relative to the sector and market. Alpha Model Score Valuation Growth Profitability Earnings Momentum Price Momentum Quality Capital & Funding Liquidity Risk Technicals & Trading Normalized Score Percentile relative to sector(/1032) Percentile relative to market(/423) Source (all charts): FactSet, Thomson Reuters, and Macquarie Research. For more details on the Macquarie Alpha model or for more customised analysis and screens, please contact the Macquarie Global Quantitative/Custom Products Group (cpg@macquarie.com) 15 February

20 Important disclosures: Recommendation definitions Macquarie - Australia/New Zealand Outperform return >3% in excess of benchmark return Neutral return within 3% of benchmark return Underperform return >3% below benchmark return Benchmark return is determined by long term nominal GDP growth plus 12 month forward market dividend yield Macquarie Asia/Europe Outperform expected return >+10% Neutral expected return from -10% to +10% Underperform expected return <-10% Macquarie South Africa Outperform expected return >+10% Neutral expected return from -10% to +10% Underperform expected return <-10% Macquarie - Canada Outperform return >5% in excess of benchmark return Neutral return within 5% of benchmark return Underperform return >5% below benchmark return Macquarie - USA Outperform (Buy) return >5% in excess of Russell 3000 index return Neutral (Hold) return within 5% of Russell 3000 index return Underperform (Sell) return >5% below Russell 3000 index return Volatility index definition* This is calculated from the volatility of historical price movements. Very high highest risk Stock should be expected to move up or down % in a year investors should be aware this stock is highly speculative. High stock should be expected to move up or down at least 40 60% in a year investors should be aware this stock could be speculative. Medium stock should be expected to move up or down at least 30 40% in a year. Low medium stock should be expected to move up or down at least 25 30% in a year. Low stock should be expected to move up or down at least 15 25% in a year. * Applicable to Asia/Australian/NZ/Canada stocks only Recommendations 12 months Note: Quant recommendations may differ from Fundamental Analyst recommendations Financial definitions All "Adjusted" data items have had the following adjustments made: Added back: goodwill amortisation, provision for catastrophe reserves, IFRS derivatives & hedging, IFRS impairments & IFRS interest expense Excluded: non recurring items, asset revals, property revals, appraisal value uplift, preference dividends & minority interests EPS = adjusted net profit / efpowa* ROA = adjusted ebit / average total assets ROA Banks/Insurance = adjusted net profit /average total assets ROE = adjusted net profit / average shareholders funds Gross cashflow = adjusted net profit + depreciation *equivalent fully paid ordinary weighted average number of shares All Reported numbers for Australian/NZ listed stocks are modelled under IFRS (International Financial Reporting Standards). Recommendation proportions For quarter ending 31 December 2016 AU/NZ Asia RSA USA CA EUR Outperform 57.53% 50.72% 45.57% 42.28% 60.58% 52.79% (for global coverage by Macquarie, 8.71% of stocks followed are investment banking clients) Neutral 33.90% 33.97% 43.04% 50.11% 37.23% 35.62% (for global coverage by Macquarie, 8.05% of stocks followed are investment banking clients) Underperform 8.56% 15.30% 11.39% 7.61% 2.19% 11.59% (for global coverage by Macquarie, 4.63% of stocks followed are investment banking clients) GPT AU vs ASX 200 Prop, & rec history SCG AU vs ASX 200 Prop, & rec history VCX AU vs ASX 200 Prop, & rec history (all figures in AUD currency unless noted) (all figures in AUD currency unless noted) (all figures in AUD currency unless noted) Note: Recommendation timeline if not a continuous line, then there was no Macquarie coverage at the time or there was an embargo period. Source: FactSet, Macquarie Research, February month target price methodology GPT AU: A$5.09 based on a NAV methodology SCG AU: A$4.51 based on a NAV methodology VCX AU: A$2.93 based on a NAV methodology Company-specific disclosures: GPT AU: Macquarie and its affiliates collectively and beneficially own or control 1% or more of any class of 's equity securities. SCG AU: Macquarie Group Limited together with its affiliates beneficially owns 1% or more of the equity securities of Scentre Group Ltd. VCX AU: MACQUARIE CAPITAL (AUSTRALIA) LIMITED or one of its affiliates has provided Federation Centres Ltd with investment advisory services in the past 24 months, for which it received compensation. Important disclosure information regarding the subject companies covered in this report is available at Date Stock Code (BBG code) Recommendation Target Price 15-Aug-2016 GPT AU Underperform A$ Jul-2016 GPT AU Neutral A$ Apr-2016 GPT AU Neutral A$ Mar-2016 GPT AU Neutral A$ Jan-2016 GPT AU Outperform A$ Aug-2015 GPT AU Outperform A$ Feb-2015 GPT AU Outperform A$ Jan-2015 GPT AU Outperform A$ Jan-2015 GPT AU Outperform A$ Oct-2014 GPT AU Outperform A$ Oct-2014 GPT AU Outperform A$ Aug-2014 GPT AU Outperform A$ Feb-2014 GPT AU Outperform A$ Jan-2014 GPT AU Outperform A$ February