Interesting work on TOWR. My question is how involved is the Indonesian government in encouraging tower sharing? And do you think the ROICs on lower tenancy ratios still strong?
Hi, thanks for the question! Yes Indo government is quite involved in the tower business. At the national level, you have the Joint Regulation of 2009 (signed across four ministries) that explicitly provides guidelines for joint use of telecom towers. In the 2021 Omnibus Law implementing regulation (GR 46/2021) went further — it shifted the language from encouraging operators to share passive infrastructure to obligating them to open access.
But this regulatory support is a policy choice, not a constitutional guarantee. A future administration could still reprioritize.
On ROIC, management shares that it is at 11.4% on a single-tenant tower, which sounds decent. But I'd still be cautious because the spread over TOWR's cost of capital especially an environment where Indonesian interest rates have been elevated (BI rate was at 6% for a stretch) and TOWR carries meaningful debt to build or acquire these towers for tenancy. Which is why as an investor, higher tenancy ratio is preferred (lower marginal cost of adding a new tenant at the same tower).
Yeah I think it's an interesting scenario if the growth if good and tenancies will rise. I looked at Vietnam telco tower business and the problem was the towerco was owned byt he telco so you have a conflict of interest in pricing and the regulation wasn't strict on tower sharing. But overall towers is a great business if theres NIMBY for competitors to set up towers near yours.
Hi Toby great writing as always! Just read Marks notes on AI hurtles Ahead :). QQ I'm starting to use Claude for researching companies for investing, but haven't tried NotebookLM. Any comparison highlight worth mentioning for these two based on your experience?
Claude is where I'd go if I want to build stuff. Simulations, app, statistical analysis etc. It is also quite critical that can help you challenge your assumptions as well. But of course you also need to be alert of latent assumptions it can inevitably make.
NotebookLM is where I'd go when I need to interrogate the source material. You feed it the actual annual reports and filings, and it stays grounded in those documents (doesn't really halucinate). It has a convenient UI where you can click on the citations and be pointed directly to the source it quotes, that you provide. It is also great at making mindmaps and reports to help me study better.
Interesting work on TOWR. My question is how involved is the Indonesian government in encouraging tower sharing? And do you think the ROICs on lower tenancy ratios still strong?
Hi, thanks for the question! Yes Indo government is quite involved in the tower business. At the national level, you have the Joint Regulation of 2009 (signed across four ministries) that explicitly provides guidelines for joint use of telecom towers. In the 2021 Omnibus Law implementing regulation (GR 46/2021) went further — it shifted the language from encouraging operators to share passive infrastructure to obligating them to open access.
But this regulatory support is a policy choice, not a constitutional guarantee. A future administration could still reprioritize.
On ROIC, management shares that it is at 11.4% on a single-tenant tower, which sounds decent. But I'd still be cautious because the spread over TOWR's cost of capital especially an environment where Indonesian interest rates have been elevated (BI rate was at 6% for a stretch) and TOWR carries meaningful debt to build or acquire these towers for tenancy. Which is why as an investor, higher tenancy ratio is preferred (lower marginal cost of adding a new tenant at the same tower).
Hope this helps! :)
Yeah I think it's an interesting scenario if the growth if good and tenancies will rise. I looked at Vietnam telco tower business and the problem was the towerco was owned byt he telco so you have a conflict of interest in pricing and the regulation wasn't strict on tower sharing. But overall towers is a great business if theres NIMBY for competitors to set up towers near yours.
Hi Toby great writing as always! Just read Marks notes on AI hurtles Ahead :). QQ I'm starting to use Claude for researching companies for investing, but haven't tried NotebookLM. Any comparison highlight worth mentioning for these two based on your experience?
Thank you for your kind compliment!
Claude is where I'd go if I want to build stuff. Simulations, app, statistical analysis etc. It is also quite critical that can help you challenge your assumptions as well. But of course you also need to be alert of latent assumptions it can inevitably make.
NotebookLM is where I'd go when I need to interrogate the source material. You feed it the actual annual reports and filings, and it stays grounded in those documents (doesn't really halucinate). It has a convenient UI where you can click on the citations and be pointed directly to the source it quotes, that you provide. It is also great at making mindmaps and reports to help me study better.