The Nuts & Bolts of Infrastructure ETFs


Barron’s had an article today about the infrastructure plan that was discussed regularly during the 2016 presidential campaign but that doesn’t seem to have been given much consideration since. The investment angle in the article said it would help certain industrial stocks including Vulcan Materials (VMC), Fluor (FLR) and a few more.

I have been bullish on the theme for most of that time over the last dozen or so years on the simple premise that the money is going to be spent. It may be spent inefficiently or efficiently, it may be fits and starts but there is no question that on a global basis this has been and will continue to happen. That creates a tailwind that certainly doesn’t guarantee anything but doesn’t make it worse either.

Infrastructure ETFs have been around a little over ten years. There are also other exchange traded vehicles that offer access to various niches within the space. Infrastructure is a great example where many funds in the “same space” offer very different exposures creating the need to look closely under the hood.

The first ETF was the SPDR S&P Global Infrastructure ETF (GII), it is also one of the biggest along with iShares Global Infrastructure (IGF) and although newer, the FlexShares STOXX Global Broad Infrastructure Index Fund (NFRA) is also pretty big in terms of AUM.

At the sector level, infrastructure can incorporate a lot of different things. When GII first listed, it had 90% in utilities and at the time I said it would simply (and obviously) be a proxy for the utility sector. Subsequently the underlying index was changed and how it has about 40% in utilities and 40% in industrials. This is similar to IGF which has 40% in utilities and 40% in transportation which is a subset of industrials. The funds have traded similarly in recent years without offering much to write home about in terms of performance. NFRA oddly (to me) has its largest sector weighting to telecommunications, actually tied with energy, each at 29%.

While I bet there is a persuasive argument to be made for telecom as infrastructure I would not want Verizon (VZ) and AT&T (T) as my two largest positions at about 4% each in the fund to express that argument followed by 3.5% more in Comcast (CMCSA). cites 11 funds in the infrastructure niche and most of them offer broad or broad-ish sector diversification except for maybe the First Trust Nasdaq Clean Edge Smart GRID Infrastructure Index (GRID) which appears to tilt more heavily to tech and industrials.

Using GRID or a fund like Global X US Infrastructure Development ETF (PAVE) as a proxy for industrials, which have a 60% weight in the fund can make sense for someone who buys into the above mentioned Barron’s article but something that is 40% in utilities, PAVE has 3% in utilities, wouldn’t make as much sense as the low volatility of utilities could easily offset the industrial exposure. Underscoring that point, GII has a much closer correlation to utilities than it does to industrials.

One thing to consider with any narrow based theme is that you might be better off with individual stocks and that might be the case with infrastructure. I believe toll roads are the most interesting part of infrastructure but there is no toll road ETF. The ProShares fund with symbol TOLZ only has 6% in toll roads. I view them as low volatility, cash flow heavy, dividend payers which would essentially be the opposite of what Barron’s is talking about this week. I’ve written many times about these in the past. The extent to which they deliver on the low volatility part admittedly ebbs and flows but does so enough for me to have faith in the exposure.

Airports could also be lower beta dividend payers but like most of the toll roads, these would be harder to trade foreign stocks. A couple of examples; Sydney Airport yields 4%, Auckland Airport yields 3.71%. I will write more about these later.

I don’t currently have any exposure for clients to any infrastructure ETFs. I think of publicly traded exchanges as being infrastructure-ish and have owned CBOE for clients for quite a few years. Looking forward, payment infrastructure could be thought of also being infrastructure-ish including the credit cards, most client own Master Card (MA), most clients also own Square (SQ) which is a fit here too.