Now that the Bill of Quantities (BoQ) of the controversial ZTE National Broadband Network Project by the Philippine government has been released, I was asked to take a look and see what I make out of it. Here are my findings so far.
The Bill of Quantities is an annex of documents which outline the details of the project costing from the equipments to the services and trainings that will be undergone to implement the project. By checking and analyzing some of the data in there, we can make reasonable conclusions if the project were indeed padded or not.
First stop, the topology. This will illustrate how the network will be interconnected and with what. I’ve drawn a diagram below as best as I could to show how they plan to do it.
The primary line of transmission is via the NBN Backbone distributed thru microwave relay stations deployed nationwide. They will reach majority of the WiMax Base Stations and transmitted to the nodes (regional government offices, municipalities and barangays).
The secondary transmission is done via commercial networks — meaning, if the microwave relay stations cannot reach/cover a certain base station, they will hook them up via 2Mbps ADSL from local Internet Service Providers (ISPs).
The 3rd mode of transmission is via VSAT (Very Small Aperture Terminal, via Satellite Internet) which can get access up to 4Mbps to the WiMax base stations. This is a back-up plan for locations where there are no local/commercial broadband service and relay stations cannot reach. They’ve allocated 5% (or 15 of the 300 base stations) of the total base stations that might avail of this. This mode is the most expensive of the 3 in terms of monthly link charges.
Next, let’s look at the breakdown of the costing. The total contract price (TCP) of the project is pegged at $329,281,290 or about Php13.5 Billion at current exchange rates. This is broken down into two — Equipment Sales of $194,051,628 and Service Charges of $135,429,662.
The Equipment cost is further distributed as follows:
Transmission Backbone Equipment: $47,649,037
Regional WiMax System: $82,077,545
IP Backbone Equipment: $15,510,251
VoIP System: $15,175,887
DC/NOC Equipment: $8,977,882
Information Security system: $6,743,012
Monitor & Environment Equipment: $3,532,877
Uniform NMS: $2,143,540
DC Power & Diesel Generator: $11,505,187
Total Price: $194,051,628
It’s very hard to determine if these equipments were overpriced. Some of them, you can’t even search on the internet — like those Microwave transmitters carrying 4 x E1 to 16 x E1 connections which costs between $1.2 to $4.5M each according to the BoQ. So, I looked at a few samples which we can obviously compare to current market prices.
Office Printer : 1 set @ $804.20
Desktop Computer : 1 set @ $2,055.94
Laptop Computer : 1 set @ $3,440.56
Basic E1 Tester : 5 units @ $7,128.41
MS SQL Server 2005 Standard Ed. : 1 CPU License @ $6,797.73
These items, though a bit pricey, are still within
tolerable reasonable levels considering 2007 market prices. I searched on eBay for an E1 Tester and saw one at $3,500 and the current 1 CPU license for MS SQL Server is $5,999 according to the Microsoft website.
Then, I looked at connection/link charges that were included in the pricing:
Manila IDC: 100Mbps/155M SDH Link = $30,000/month
Regional PRI to PSTN: E1 $500 x 43 = $21,500/month
Regional Sites DSL: 2MB/128K CIR @ $80 x 300 = $24,000/month
VSAT: E1 (2Mbps) $7,000 x 15 = $105,000/month (est. 5% of total)
That brings the annual connection cost of $2,166,000. The ZTE contract price includes this for up to first 3 years only ($6.5M). That means after that, the government will have to shoulder it back again. Commercial price of an E1 connection is at $500 and the $80 for a 2Mbps DSL with 128Kb CIR (committed internet rate) seemed normal. The VSAT is obviously very expensive considering it will probably only cover 5% of the total base stations.
So far so good. Now, we look at the breakdown of the Services cost:
Site Prep & Civil Works: $48,571,040
System Implementation & Engineering: $20,607,490
Site Engineering for Remote Office Site: $28,686,840
Service for IDC Construction & Integration: $8,287,281
Joint Project Management Office: $5,995,000
Link Charges: $6,498,000
Managed Services: $14,875,507
Total Price: $135,429,662
The problem with service costing is that there are no standards and they’re mostly priced arbitrarily. I find it rather interesting that for a huge infrastructural project such as this, services take up 41.1% of the total cost. Take for example the training cost ($1.95M) — it will take $1.3 M to bring 20 people to China to train how to run the Microwave/WiMax equipments. Each course costs an average $12,000 per person. Likewise, they will spend another $600,000 to train 40 people locally on the same courses. I thought if you buy some sophisticated equipment, the vendor will train someone from the buyer how to operate it for free?
So, what’s the verdict? Let me qualify this by answering specific questions.
Is the ZTE proposal a sound strategy for a national broadband plan? IMO, it looks like it. They’ve maximized the WiMax implementation, coupled with local ISP connectivity and a VSAT backup. I think this approach is practical and cost effective in itself.
Does the NBN Project really cover 100% of the country as it claimed it can? Yes, it can — the VSAT backup practically guarantees that.
Is there overpricing in the hardware/equipment costs? Nothing too obvious nor significant — my sample comparison of some listed items with market prices turned out reasonable.
Is there overpricing in the services component? I cannot categorically say ‘yes’, but there’s some semblance of it. Besides, service charging is mostly a guesstimate so you can fairly say that they’ve somehow padded these prices a bit. If by how much, I cannot say. Only way to calibrate that is by asking 2 other suppliers to send in their costing for the same project.
So, where’s the catch? Here’s my theory. This is not your ordinary “where’s my cut” type of underhanded deal where suppliers bid for government projects with overpriced materials like a Php20,000 toilet seat or Php1,500 for an ordinary broom. That’s so amateurish.
For huge projects like this, the “payouts” are drawn from wholesale discounts. So, instead of say a 30% discount to nab the deal, they’ll charge in full and re-allocate the 30% corporate discount to the pockets of those who will guarantee the approval of the project. Incidentally, a 30% discount amounts to $100M so it’s not a bad deal after all, if that were the case. That way, the Bill of Quantities will still pass thru rigid scrutiny.
Note: There are so many other stuff I haven’t covered here, esp. the details. You may download a copy of the Bill of Quantities here so you can see it for yourself and make your own personal assessment.