Exchange-traded fund (ETF) total expense ratios (TERs) are witnessing price wars locally and in the US.
Recently Satrix dropped the TER in their flagship STX40* to 0.1% while the CoreShares CTOP50 currently has the lowest TER in South Africa at 0.07%.
I remember being charged some 6% for a unit trust investment in the late 1990s, although this did include adviser and platform fees and no performance fees (as it never performed) – and I have never bought a unit trust since.
More recently, a price below 0.4% was considered exciting for an ETF. The more generic ETFs are involved in price wars.
This is great for the consumer – but are we going to see a zero-fee ETF locally or globally?
An article by ETF Database showed that the 74 cheapest ETFs listed in the US have TERs of 0.07% or lower, with five coming in at 0.03%.
At the start of 2017, 0.05% was considered cheap in the US, and while 0.02 percentage point is not massive, it means more money in the client’s pocket, and it will add up over the years.
The thing is that with a generic ETF issued over, say, the S&P500, or locally the Top40, the only real difference between them is the fee being charged. This raises the question of how low ETF fees can go.
Frankly, it only seems a matter of time until a zero-fee ETF is launched in the US, and I think they'll have such a product within the next few months.
After that we may see negative-fee ETFs, in other words you get paid to hold the ETF. That may sound crazy, but ETFs are a trillion-dollar business, and competition for market share is fierce.
Locally we’re seeing downward pressure on TERs, but zero fees may be some way off. It is also important to note that these very low TERs are on the simple generic ETFs, not exotic underlying ETFs with low TERs.
A TER of zero may become a reality in a number of ways. First, it could be a loss leader, drawing investors into a platform that may encourage clients to buy other products that are not zero-fee.
Second, the ETF issuer could use script- lending as a way to generate income, offsetting their costs and making a modest profit.
Script-lending is when long-term holders of a share (such as an ETF issuer) lend some of the shares to those wanting to short the shares.
Shorting is selling what you don’t own, and you must have the shares to sell them, so the short seller would borrow them, paying a small fee to the owner.
The lender also has to pay any dividends due – so as ETF holders we’ll be no worse off, and the issuer will generate some revenue.
Way back in the early days of STX40, the TER was often either very small or zero due to the revenue from lending script to short-sellers.
The race to zero fees is because it is very difficult for issuers to stand out in crowded markets.
For example, locally we have four ETFs issued over the Top40 (Satrix, Ashburton, Stanlib and, most recently, Sygnia) and three issued on the S&P500.
They all track exactly the same index, and they need to differentiate themselves. This can be via a platform – but price is the only real differentiator.
A last important point is that locally issued offshore and exotic ETFs will likely see some price drops, but they’re a very long way from zero due to the extra costs involved.
Exotic ETFs have to pay the index maker while offshore ones feature extra cost layers due to currency conversions.
This article originally appeared in the 16 November edition of finweek. Buy and download the magazine here.