Keppel DC REIT (SGX: AJBU) analysis (Updated 15 Dec 2020)

On the back of my previous analysis post about Mapletree Industrial Trust, I decided to follow up with another analysis post. After all, as the saying goes, consistency breeds disciplinešŸ˜….

In this post, I will be looking at Keppel DC REIT (KDCR).

In recent times, KDCR has garnered significant interest from retail investors. This is mainly due to the digitalization process that many companies are currently undergoing, leading to a huge demand for online data storage.

Thus, it is no surprise that KDCR’s share price has been on an upward trend since its IPO.

While there has been some pullback in its stock price recently, IPO investors would still be looking at returns of at least 200% (dividends included) at the current market price of $2.70.

Given KDCR’s impressive performance since its IPO, I thus decided to analyze and see if it is a good addition to my portfolio.

Read on to find out my review of KDCR’s performance for the past 5 years, which also include results from 3Q 2020.

Disclaimer: This post is mainly for education and sharing purposes. It should not be seen as a recommendation to buy or sell a stock. Such decisions should be made on the reader’s own personal accord, after doing their due research. As such, I take no responsibility for any future loss or gain resulting from the purchase or sell of this stock. I also do not currently own any shares in KDCR.

Business Overview

As a REIT, KDCR derives its income from the leasing and management of data centres.

There are 3 types of data centres in KDCR’s portfolio, namely colocation, fully-fitted and shell and core.

  • Colocation data centres – Colocation data centres are typically leased out to multiple clients. Such leases comprise of both short-term retail leases or longer-term wholesale lease to bigger companies.
  • Fully-fitted data centres – These data centres are ones which are fully equipped to deal with all forms of network demands. For KDCR, most of its fully-fitted data centres are purpose-built for particular clients.
  • Shell and core – Similarly to fully-fitted data centres, KDCR’s shell and core are also purpose-built for clients. The difference is that shell and core data centres only contain the skeletal requirements to run a data centre. Thus, the mechanical and electrical infrastructure required for extensive networking demands is not provided by KDCR.

Currently, KDCR has a portfolio of 19 data centres, with the majority being colocation data centres. As such, it is unsurprising that colocation data centres contributed to the majority of KDCR’s revenue in FY 2019.

As seen, around 75% of KDCR’s revenue in FY 2019 came from its colocation data centres. This means that KDCR’s revenue can be subjected to some volatility due to the shorter lease nature for colocation data centres. However, based on its results for the past 5 years, this has had little impact.

Having had a brief understanding of what KDCR is about, I will now move on to the good parts about it.

What I like about KDCR

Impressive top and bottom line growth

Looking at its results for the past 5 years, KDCR has really impressed me with its top and bottom-line performance.

As seen, gross revenue, net property income and distributable income all recorded compounded annual growth rate (CAGR) of double digits. This trend looks set to continue in FY 2020 given the current figures to date for 3Q 2020.

If one was to dissect KDCR’s revenue growth, we can see that much of the growth came from its colocation data centres.

Although shell and core data centres grew at a much higher rate, they constituent too small a portion of KDCR’s total revenue to have a huge influence on revenue growth.

Another contributing factor to KDCR’s strong top and bottom-line performance is its relative constant property-related expenses over the years.

KDCR property-related expenses

Over the past 5 years, KDCR’s property-related expenses have not really increased that much. Much of this is down to triple and double-net leases for KDCR’s fully-fitted and shell and core data centres. Under such leases, the client, instead of KDCR, bears all or a majority of property-related expenses.

Although property-related expenses would most likely increase in FY 2020, I strongly believe that KDCR would be able to continue its double-digit growth for FY 2020.

Positive returns for unitholders

Besides having double digit growth in its top and bottom line, KDCR has also steadily provided positive returns for unitholders.

Bar a slight dip in DPU for FY 2016 due to a rights issue, we can see that there have been positive year-on-year growth in KDCR’s DPU and NAV. Furthermore, this positive momentum looks set to continue for FY 2020.

In addition to growth in DPU and NAV, AFFO per unit also grew at a CAGR of 5%. This means that more net cash from operations is being generated per unit of KDCR.

With the growth in DPU, NAV and AFFO per unit, unitholders can be assured that they are owning a REIT that both increases in value and dividends paid out over time.

Sustainable debt profile

In addition to its impressive financial results, KDCR also score high points in terms of debt management.

In terms of interest coverage, KDCR is amongst one of the highest in the S-REITs market with a coverage of 12.7 times. This means that bar any exceptional circumstances, KDCR should have no problems fulfilling its debt repayments.

From the image above, we can also see that KDCR’s cost of debt has been decreasing over the years. This is good news as a lower cost of debt equates to higher overall financial savings.

Coupled with a decreasing cost of debt, KDCR’s current aggregate leverage also affords it ample debt headroom before hitting the current regulatory limit of 50%. KDCR’s debt tenor also looks healthy at 3.4 years as of 3Q 2020.

All in all, KDCR’s current debt profile allows it ample room and opportunities to grow as a REIT.

High portfolio occupancy and long WALE

The last aspect of KDCR which I like has got to be its portfolio occupancy and weighted average lease expiry (WALE).

As seen, KDCR has a really healthy portfolio occupancy of 96.7%. In addition, KDCR has also experienced year-on-year increases in its portfolio occupancy since FY 2017.

If we were to focus purely on colocation data centres, occupancy rates also look quite healthy.

Out of KDCR’s 9 colocation data centres, only 3 were below 90% occupancy. Of these three, Keppel DC Singapore 5 is still currently undergoing asset enhancement works while Keppel DC Dublin 1 only finished practical asset enhancement works in 1H 2020. Thus, more run time is needed to evaluate the occupancy performance for these two data centres.

The only laggard data centre would be Basis Bay Data Centre. However, this data centre currently only makes up 1% of KDCR’s AUM as of FY 2019. Hence, even with its relative low occupancy rate, it does not have a significant impact on KDCR’s revenue and earnings.

In terms of KDCR’s WALE, it stood at 7.2 years for Sep 2020.

Much of KDCR’s WALE can be attributed to leases from its fully-fitted and shell and core data centres.

While WALE for colocation data centres has been decreasing, I’m not too worried about it. This is largely due to KDCR’s relatively healthy lease expiry profile.

Source: KDCR’s 3Q 2020 operational updates

As seen, only around 16.1% of leased area is up for renewal over the next two years. Given that data centre clients are often sticky in nature due to the required investment cost in servers and racks, KDCR should not have issues in negotiating renewals for most of its clients.

In the worst-case scenario that none of these clients renewed their leases, KDCR should have no difficulties in attracting new clients given the continual demand for data centre space.

Points to consider for KDCR

While there is much to like about KDCR, I also came across a few aspects which warrant a second look. I would not consider these aspects as negative per se, rather they should be considered as potential risks instead.

Inconsistent net profit growth

While KDCR’s growth in gross revenue, net property income and distributable income has been impressive so far, this has not been replicated in its net profit growth.

As seen, KDCR’s net profit has been up and down for the past 5 years. Much of this is due to net value losses on KDCR’s investment properties and losses in foreign currency translation movement.

While both these values are non-cash items, meaning they do not affect the amount of dividends KDCR pays out, it can still affect the REIT in the long run. This is especially true for losses on investment properties.

With regards to net value loss on investment properties, it can affect the amount of negotiated rent between KDCR and potential clients. After all, lesser valued properties can result in lower rental rates. Hence, if a REIT experiences continual net value losses on its investment properties, there might be an adverse impact on its future rental income.

High payout ratio

When it comes to dividend payout ratio for REITs, I always like to compare it to the REIT’s AFFO. Doing so allows me to see if the dividends can be sustained by the REIT’s operations or not.

In the case of KDCR, there were instances where the dividends surpassed that of its AFFO.

As seen, dividends in FY 2015 and FY 2018 surpassed AFFO figures. For the other years, dividend payout ratio over AFFO ranged between 70% to mid 80s%.

Having said that, I do not think that unitholders should be too worried about KDCR’s dividend payout ratio. This is mainly due to the fact that its dividend payout ratio has not consistently exceeded its AFFO. Furthermore, as a REIT, KDCR is required to pay out a large portion of its taxable income as dividends.

Nevertheless, unitholders should still monitor this figure in the years to come.

Near-term refinancing obligations

When it comes to KDCR’s debt maturity profile, it is definitely not very well spread out.

Source: KDCR’s 3Q 2020 operational updates

As seen, around 29.3% of KDCR’s current debt matures in the next 2 years, 12% in 2021 and 17.3% in 2022. In total, this works out to around 316.4 Million based on KDCR’s current debt of 1.08 Billion.

Having said that, based on its balance sheet in 1H 2020, I do not foresee there being any issues in KDCR refinancing or paying off the loan that would be due next year. Furthermore, KDCR can also take on more debt given its current aggregate leverage percentage. Doing so would potentially better spread out the debt profile.

If that happens, unitholders should expect the aggregate leverage percentage to inch up slightly if KDCR takes on more debt. NAV might also decrease should cash be used or units are issued to pay off the debt.

Nevertheless, these remain as mere speculation. Only time will tell what KDCR does to refinance its upcoming debt.

Concentration risk

The last aspect to note for KDCR is its apparent concentration risk in rental income and revenue.

As seen, much of KDCR’s rental income comes from the internet enterprise sector. This is also true with regards to the top ten rental contributors whereby one client from the internet enterprise sector contributed to a massive 41% of KDCR’s rental income in 2019.

While having internet enterprise as a significant sector is not surprising given the required networking demands, I was slightly taken aback that one client has such a significant influence on KDCR’s rental income. This is extremely risky as a huge void needs to be filled should that one client choose not to renew its lease with KDCR.

In terms of geographical revenue, Singapore continues to be the main contributor to KDCR’s revenue. In fact, the revenue contribution from Singapore has significantly increased over the past few years.

As seen, Singapore’s revenue contribution has the highest CAGR over the past 5 years as compared to other regions. Thus, it is fair to say that the state of Singapore’s data centre market would continue to have significant influence in KDCR’s revenue trajectory.

Future outlook

When it comes to determining the future outlook of KDCR, I feel that there are three main aspects to be looked at. These are the potential growth of the data centre industry, the attractiveness of Singapore as a data centre market and the strength of KDCR’s sponsor.

The data centre industry

As a whole, the data centre industry should see immense growth in the coming years. This is mainly due to the acceleration of the economic digitalization and the ongoing implementation of 5G.

This bodes well for KDCR as there would be increased demand for colocation and purpose-built data centres. In fact, I see immense growth potential in the colocation data centre space for KDCR.

This is due to the immense cost needed in maintaining on-premise data centres. As such, most companies, especially SMEs, would look to cloud providers to cut costs instead of independently maintaining their own network equipment or data centres. Incidentally, cloud providers often form the bulk of clients in colocation data centres. While KDCR does not openly disclose their clients, I expect it to be pretty much similar to global trends.

Hence with the continual rise in data demands, I expect KDCR to continue doing well in the near future given the significance of colocation data centres in its revenue.

The Singapore data centre market

Given that more than 50% of KDCR’s revenue is derived from Singapore, assessing the prospect of Singapore’s data centre market is crucial to predicting KDCR’s outlook in the near future.

Fortunately, Singapore continues to be a prominent data centre player in the Southeast Asian region. This is mainly due to Singapore being a relatively safe country with a highly interconnected population. However, growth in Singapore’s data centre market might just be reaching a saturation point soon.

Land space and sustainability

It is no secret that Singapore is really small as compared to our Southeast Asia neighbours. Hence, we cannot afford to continue building new data centres even if Singapore remains an attractive location to industry players. In addition, there is also the issue of sustainability to consider.

Given that data centres rely on a huge amount of power to function, building more data centres would fly in the face of Singapore”s commitment to the Paris Climate Accord. Thus, it was not surprising when the government hit the pause button in the building of new data centres.

In terms of improving sustainability, KDCR has taken steps such as using renewable energy sources and water recirculation technology for its data centres. However, the issue of land is one that is much harder to circumvent.

While technologies such as floating or underwater data centres are being looked at, it would take some time for these initiatives to be fully implemented.

Thus, in the short term, the Singapore market would be a double-edged sword for KDCR. On one hand, rental rates for Singapore data centres might increase due to the limited supply. On the other, there would be lesser growth opportunities in Singapore for KDCR.

Sponsor’s strength

With the potential dearth of new data centres in Singapore, KDCR has to rely on overseas opportunities to grow. This is where its sponsor has a huge role to play.

If you did not already know, KDCR’s sponsor is Keppel Telecommunications and Transportation (Keppel T&T). Based on Keppel T&T’s potential pipeline of properties, it does provide certain opportunities for KDCR.

As seen, most of the potential data centres from its sponsor are in mature markets, with the exception of Malaysia. Furthermore, these data centres could also potentially steer KDCR away from a predominantly Singapore core in its revenue. This would only bode well for the REIT in the long run.

However, with only three potential mature properties for acquisition, I think it is fair to say that Keppel T&T would not have significance influence in KDCR’s opportunity to grow. As such, KDCR has to source out its own growth opportunities instead of relying on its sponsor.

For now, I feel that KDCR’s short term outlook remains positive. In the long term though, I would be mindful of how KDCR branches away from its reliance on Singapore to pursue overseas opportunities. Whether it would be a success or not can be witnessed in KDCR’s future financial figures.


When it comes to a REIT’s valuation, I like to depend on its PB ratio, AFFO yield and distribution yield.

Price-to-book ratio

At a market price of $2.70, KDCR’s PB ratio stands at 2.33. To determine if it is a fair valuation, I compared it to KDCR’s historical PB ratio.

As seen, the 5-year PB ratio average for KDCR is around 1.62, which translates to a market price of $1.88. This represents a 30.4% drop from the current market price.

For KDCR to hit the price of $1.88, it would require a market correction of sorts seen in March 2020. Given the optimism surrounding the stock market of late, I do not expect KDCR’s price to fall to $1.88 any time soon. Hence, investors looking to invest in KDCR based upon a fair PB ratio might have to wait till the next market correction.

AFFO yield

At its current market price, KDCR’s AFFO yield stands at 2.77%. While I often compare the current AFFO yield to the REIT’s cap rate, doing so for KDCR would be unrealistic given the bullish market sentiments about the REIT in recent years.

As of FY 2019, KDCR’s internal cap rate (net property income/value of investment properties) stood at 6.72%. For KDCR’s AFFO yield to hit that level, the market price has to decrease till $1.11. I just do not see that happening, market correction or not.

As such, I thus decided to look at KDCR’s historical AFFO yield as a comparison instead.

As seen, KDCR’s AFFO yield has been decreasing over the years. This means that the market is placing a premium valuation on KDCR perhaps due to its stellar performance over the years.

Using FY 2019’s AFFO yield and incorporating a 5% growth rate of AFFO per unit in 2020, a fair entry price would thus be $2.15.

Distribution yield

Based on the current market price, KDCR’s distribution yield stands at 2.62%. KDCR’s own internal distribution yield (distributable income/value of investment properties) stood at 4.30%.

As such, for the distribution yield to increase to 4.3%, KDCR’s market price has to be around $1.64. Whether it would drop to such levels in the next market correction remains to be seen.

From the three financial metrics above, we can roughly conclude that the current market price for KDCR is a tad overpriced. For potential investors, a good entry price I feel would range from $1.64 to $2.15. On which spectrum of the range you would buy KDCR at would be dependent on your own risk appetite and perspective of KDCR’s potential growth.

Personally, I would purchase KDCR at around $2. Hence, I would patiently wait for the next market correction before investing in KDCR.

Concluding thoughts

Overall, I feel that KDCR is a solid REIT to invest in. While its growth prospects in the local scene might be subdued, there should be an abundance of overseas opportunities given the global growth of the data centre market.

Furthermore, a solid financial position also enables KDCR to position itself nicely for future acquisitions. Whether such acquisitions would be DPU and NAV accreditive remains to be seen. However, I do have high hopes with how KDCR’s DPU and NAV have been over the past years.

Nevertheless, I feel that now is not the right time to invest in KDCR given its rather high market valuation. Potential investors might want to wait for the next market correction before investing so as to have a greater margin of safety.

Do you agree with my assessment of Keppel DC REIT? Let me know in the comments I would love to get some feedback!:)

Found this useful? Share it!

Similar Posts


  1. Very good and detailed write up.
    Are your FFO and AFFO formula as below as per Investopedia ?
    FFO = net income + amortization + depreciation – capital gains from property sales
    AFFO = FFO + rent increases – capital expenditures – routine maintenance amounts

    Do you mind to share how you get the “rent increases” and “routine maintenance amounts” ?

    1. Hmm, a little different.

      For my AFFO I based it off the cash flow statement, so
      AFFO = Cash generated from operations before working capital – interest paid – Capex – tax paid

      The Capex portion is on the grey side coz KDCR does not really reveal what the recurring Capex is and what it is being spent on.
      But I still include the full Capex expense as Capex is not done without reason.

      For routine maintenance costs, KDCR already includes them in the property operating expenses portion of the statement of profit or loss. So net income would have already factored those in.

      1. So your AFFO is similar to FCF. Also similar to distributable income but without considered fees taken in units. For the capex, most REITs only disclose detail breakdown in Annual Report instead of quarter report, make it quite difficult to calculate AFFO or FCF if AEI/Construction is involved.

        Looking forward to your next post.

        1. Yup in a sense yes AFFO is similar to FCF.

          However, I feel AFFO differs from distributable income as AFFO looks at the net recurring cash from the REIT’s operations.
          While distributable income includes expenses and income which may not be derived from a REIT’s day to day operations.

          I think moving forth AFFO will be tougher to calculate as REITs sometimes lump Capex and construction costs in their cash flow statement in annual reports.

          Lastly, thanks for the encouragement!

          1. There is no standardize formula for AFFO as mentioned in investopedia, only FFO is standardize and show in US REIT report. You can tune it to have apple to apple comparison with REITs. I simply use distribution divided by gross revenue to check on distribution margin. To see whether management able to manage expenses effectively.

            Good post must share, Iā€™ve again shared in my Facebook group ā€“ REIT Investing Community again. Hopefully can bring some traffic to u.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.