I’m a long-term fan of investing in infrastructure funds, especially the listed ones which have become insanely popular on the LSE. I’ve invested in a great many of them over the last five years but in the last couple of years, I’ve quietly started taking some profits, though I retain a core holding of funds.

This quiet profit taking is not driven by anything other than caution – it strikes me that with many funds yielding closer to 4% (rather than 6% when I started) infrastructure is looking a tad expensive. Yet I freely admit that my caution might be misplaced, as ever more capital floods into the space, pushing down yields and pushing up the premiums on most funds.

The $64 billion question, I suppose, is what happens next? Over the next few month’s I’ll be catching up with a variety of managers who operate in this space but this week I’m going to kick off with Stephen Ellis at Gravis Capital Partners.

His flagship fund GCP Infrastructure Investments is a little different from its peers:

• Its yield is much higher at around 6% compared to 4.5% for most of its peers
• It invests primarily in senior bonds and loans (65%) unlike its more equity focused peers
• It’s also pioneered extending the infrastructure envelope to include rooftop solar projects (19% of holdings), social housing (supported living) 17%, and anaerobic digestion projects (at 16%). Classic PFI and PPP is down at 24%. This gives GCP a very different profile from its much bigger peers INPP and HICL.

GCP isn’t entirely one of a kind though – other funds have moved into its favoured niches, and like most of its peers, its shares (around 127p) currently trade at a chunky 14% premium to NAV.  Over the last year, the share price has increased 12% while over the last three years that gain is closer to 33%.

Nevertheless, Ellis is probably the right person to talk to about why many are becoming a little cautious about what actually defines infrastructure as an asset class. Should renewables or social housing really sit within the classic PFI influenced infrastructure definition? Perhaps, more importantly, should investors be worried about valuations especially given the possibility that we may all face a Corbyn led government in the next few years if the Brexit negotiations don’t go to plan – or the DUP backs out.

Overall I’d characterise Ellis’ position as cautious though optimistic. He admits that the fund is finding it “harder and harder” to originate good deals although he suggests there is more opportunity in the smaller deals end of the market. As for the diversification beyond core assets, Ellis admits that he’s finding the competition in social housing is increasing – the fund is “bidding and missing more of late and that’s largely because it has become a very attractive space yet there’s only so much demand for supported living for example”.

Looking forward he thinks the biggest current opportunity will be in new deals with the Macquarie owned Green Investment Bank. The fund has already committed £140m with the bank and the takeover by the Australian outfit should finally close within the next couple of weeks. In terms of the GIB, the biggest potential for deals will probably be in utility scale offshore wind where its already very active.

But there are also some obvious potential clouds on the horizon for infrastructure investors. Ellis says that the whole sector has grown “so far and so fast that there has been a dash for growth” and that means that the range of investments called “infrastructure” has expanded dramatically. His big concern moving forward is that there could be some notable failures which will “infect the entire space” – and “this is more than likely to happen on the fringe”. Ellis is also nervous about the obvious risks from an increasing interest rate environment – more than other funds his is a bond proxy with over 70% of assets now inflation linked. He admits that as rates increase, fund yields of around 4.5% might become a “little too tight”.

And the possibility of a Corbyn win? Ellis seems relaxed, especially when it comes to his current book and admits to not being overly worried about any hit on the pipeline of new deals – “we don’t need to grow”. But he also warns that “any incoming left leaning government is likely to be looking to tax and spend. The issue here is that they won’t be able to borrow if you are seen as walking away from your commitments”.

And talking of walking away from your commitments, what about student loans? Would GCP take the plunge and invest in the existing book? Ellis doesn’t seem enthusiastic, concerned that it is too “politically charged” and is thus not under consideration. But even ignoring the politics Ellis says he’d need to see more data history on the performance of loans before he’d even start thinking about putting money to work in the space.