Vanguard has been offering ETF classes of several of their conventional mutual shares for some time now. One of the most interesting feature, which is unique to Vanguard as far as I know, is the ability for an investor to convert shares in the mutual fund to ETF shares for a fee of $50 (free for Flagship clients if done directly at Vanguard). This conversion is one-way only — you cannot convert ETF shares into mutual fund shares.
- The main reason why an investor would want to convert mutual fund shares to the ETF class is because the expense ratios of ETF shares are lower than the investor class mutual fund shares. For example, the Vanguard emerging markets index mutual fund VEIEX charges 0.37% annually while the ETF version VWO charges only 0.25% a year.
- A second reason is the possibly better tax efficiency in terms of capital gains distributions.
- The third reason is the redemption fees savings. For certain mutual funds with redemption fees (most notably VEIEX and VTMGX), converting mutual fund shares to the ETF class and then selling the ETF shares would save a bundle on the redemption fees (0.5% for VEIEX and 1% for VTMGX shares owned for less than 5 years), especially if the amount is substamtially more than $1000.
Why not to convert?
- If an investor already owns admiral shares of the mutual fund, there is little (if any at all) saving in the expense ratio. For example, the admiral class of Vanguard’s emerging market index fund VEMAX and VWO both charges the same 0.25% expense ratio a year.
- ETFs could trade at a premium or discount, in addition to a trading spread (trading spread is the difference between the bid and asking prices). A person investing in mutual fund shares does not have to be concerned with purchasing or selling shares trading at a premium or discount, nor be concerned with the cost associated with the trading spread. Mutual funds shares are bought and solt at NAV (net asset value).
- If an investor is contributing regularly, purchasing the mutual fund “generally” has no transaction fee while purchasing an ETF would normally entail a brokerage commission charge (unless the investor is using a free-trade broker). The brokerage fees could add up quickly to a substantial amount. Note: I say “generally” because one mutual fund exception is VEIEX, which charges a 0.5% purchase fee.
The case for VEIEX to VWO conversion
After looking at the pros and cons, let’s examine the specific case for VEIEX to VWO conversion.
Expense ratio saving. If an investor is holding a substantial amount of VEIEX (say from the low four digit range onwards), there is definitely a saving when the holding is converted to VWO. The current expense ratio saving is 0.37% – 0.25% = 0.12% or 12 bps (bps = basis points) per year. This difference, compounded over many years, could become substantial.
To see if this expense ratio saving really does make a difference in the performance numbers, let’s look at the historical performance of VEIEX and VWO for the years ended Oct 31, 2006 and 2007 (values taken from the prospectus):
|VEIEX (investor shares, ER=0.42% in 2006, ER=0.37% in 2007)||32.55%||69.59%|
|VWO (ETF, ER=0.25%)||32.74%||69.78%|
|VEMAX (admiral shares, ER=0.25%)||N.A.||69.82%|
Clearly, VWO outperformed VEIEX by 0.19% in both 2006 and 2007, the first two full years of its existence.
Trading spreads. There could be a trading spread cost in selling VWO in the future during the withdrawal phase. However, from my observations, the trading volume and liquidity of VWO has been improving through the months. The impact of trading spread cost should be relatively low. Furthermore, a 12 bps expense ratio difference a year would turn into (1.0012^20)-1 = 2.43% difference in 20 years time. This should mitigate any trading spread cost.
Redemption fee. Converting the mutual fund shares to the ETF class is one way to avoid the 0.5% redemption fee.
Conversion fee. As already mentioned, Vanguard charges a $50 conversion fee unless the investor is a Flagship client. Below, we will briefly look at how this fee would affect the conversion decision.
In the analysis, assume that the $50 fee to pay for the conversion comes out from the original VEIEX investment (tax consequences not considered for simplicity). Let R be the approximate investment return a year, X be the original investment amount and N be the number of years the investment is held.
Then the value of the investment after N years, would be given by
- X*(1+R-0.0012)^N for VEIEX
- (X-50)*(1+R)^N for VWO
For VEIEX, the compounding rate is reduced by the expense ratio difference of 0.0012 a year. For VWO, the original investment amount is reduced by $50 due to the conversion fee.
For the conversion to be worthwhile, we require (X-50)*(1+R)^N to be larger than X*(1+R-0.0012)^N. For example, if R = 0.1 (10% return a year), N = 10 years, solving the inequality gives X > 4606. This means that given the assumptions, it is better to convert if the original investment amount X is at least $4606.
Other scenarios are given in the table below.
It appears that for my situation, it makes a lot of sense to convert my VEIEX holding to VWO. This is especially so given that the investment is in the mid five digit amount and the conversion will be free for me. For investors who have to pay the conversion fee, the table above may help to decide if paying the $50 fee is worthwhile.