JSE FX Trader study materials
JSE FX Trader study resources

ETF Versus Index Futures for Three-Month Exposure

JSE FX Trader · Put your knowledge to work in context.

All case studies

ETF Versus Index Futures for Three-Month Exposure

A South African investor is comparing a JSE-listed ETF tracking the FTSE/JSE Top 40 Index with a three-month JSE Top 40 futures contract. The ETF has a total expense ratio of 0.45% p.a., securities-lending revenue of 0.05% p.a., cash drag of 0.12% p.a. and a dividend withholding tax drag of 0.18% p.a. Over the past 12 months the index returned 12.00% including dividends, while the ETF's NAV returned 11.10%. The ETF currently trades at a 0.35% premium to NAV. The futures contract matures in 90 days, is priced at 8,450, while the spot index is 8,400. The three-month JSE repo rate is 8.25% and the index dividend yield is 2.80% p.a. The investor has R5 million to deploy, can earn repo on cash, and can borrow at repo plus 100 bp. The investor wants index exposure for three months and is sensitive to tracking error and financing costs.

1. Decompose the ETF's historical tracking difference and state whether the 0.35% premium changes the expected return for a buyer today.

Read the worked answer

Expected annual drag = TER 0.45% − securities-lending revenue 0.05% + cash drag 0.12% + dividend withholding tax drag 0.18% = 0.70%.

Actual tracking difference = 12.00% − 11.10% = 0.90%.

Unexplained residual = 0.90% − 0.70% = 0.20%, which could come from sampling, rebalancing timing, transaction costs or dividend timing.

For a buyer today, the 0.35% premium matters. If the premium persists, the investor merely pays more for the same NAV exposure and may recover it on sale. If it converges to zero over the holding period, the investor loses about 0.35% relative to NAV. For a three-month holding period, that is a material additional drag. The investor should not ignore the premium when comparing with futures.

2. Calculate the fair three-month futures price using cost-of-carry with the investor's cash rate and with the borrowing rate. Is the quoted 8,450 rich or cheap?

Read the worked answer

Cost-of-carry fair value:

(F = S \times [1 + (r - q)t])

With repo 8.25%: (F = 8,400 × [1 + (0.0825 − 0.0280) × 0.25]) (= 8,400 × 1.013625 = 8,514.45).

With borrowing at repo + 100 bp = 9.25%: (F = 8,400 × [1 + (0.0925 − 0.0280) × 0.25]) (= 8,400 × 1.016125 = 8,535.45).

The quoted 8,450 is below both fair values by about 64.45 and 85.45 index points, or roughly 0.76% and 1.00%. On this simple model the futures contract is cheap for a buyer. However, the investor should check dividend timing, margin financing and whether the quoted price reflects a stale spot index.

3. Advise whether to use the ETF or the futures for the three-month exposure, considering the premium, tracking difference, leverage and roll. Show a rough expected cost comparison.

Read the worked answer

Rough ETF cost over three months:

  • Annual tracking drag 0.70% × 0.25 = 0.175%.
  • Premium convergence assumption 0.35%.
  • Total expected drag ≈ 0.525%, or about R26,250 on R5 million.

Futures: the quoted 8,450 is about 64.45 points below the repo-based fair value of 8,514.45, a cheapness of roughly 0.76%. On R5 million notional that is about R38,000 of value if the contract converges normally. Against that, the investor must post margin, manage mark-to-market cash flows and roll the contract if exposure is extended. If fully cash-funded, the non-margin cash can earn repo, which is already in the fair-value calculation.

On these numbers the futures look cheaper by roughly 0.2%–0.3% over the quarter, but the advantage is not free: it depends on convergence, margin funding and operational capability. For a one-off three-month exposure with no roll, use the futures if the desk can manage margin; use the ETF if simplicity and tight operational risk matter more. Buying the ETF at a 0.35% premium is the least attractive route unless the premium is expected to persist.