Phasing small-cap companies into the

Based on recommendations from, among others, Norges Bank (see the Bank’s letter dated 20 October 2006 for a detailed discussion), and following a debate in the Storting, the Ministry of Finance decided in June 2007 to extend the composition of the fund’s strategic benchmark portfolio to include small-cap companies. These are the companies with the smallest market capitalisation in the FTSE index, as opposed to mid-cap and large-cap companies, which were already included in the fund’s benchmark portfolio.

This decision entailed a major change in the composition of the benchmark index, as small-cap companies make up a substantial proportion of the FTSE index’s universe. The transition to full exposure to the small-cap segment was carried out over a five-month period starting in October 2007 and entailed an expected investment of NOK 129 billion in more than 4 500 new companies (see Chart 7–1). The phasing-in of the new benchmark portfolio was completed at the end of the first quarter of 2008.

The previous benchmark index (FTSE All-World) is a subindex of the new benchmark index (FTSE All Cap), as the latter includes both the large-cap and mid-cap companies of the previous index and the small-cap segment. The broader All Cap index includes more than 7 000 companies, of which around 4 500 are in the small-cap segment. Together, these companies account for around 12 per cent of the new benchmark index’s total market value. Chart 7–2 shows the expected transaction requirement in the small-cap segment by country at the beginning of the transition process.

Implementation of the change

The fund’s size meant that full exposure to the additional 12 per cent of the All Cap index for which small-cap companies account would entail a large transaction volume. As liquidity in the market for small-cap companies is often limited, a large transaction volume could lead to extensive indirect transaction costs through a substantial impact on share prices. It is important to note that a change in the benchmark portfolio needs to be accompanied by a simultaneous change in the actual portfolio’s exposure in order not to increase the fund’s active risk. Two main strategies were considered when planning the transition:

a) Changing the benchmark portfolio all at once: The benchmark portfolio is changed instantly, while exposure in the actual portfolio is built up through a combination of capital inflows and sell-offs of existing investments
b) Changing the benchmark portfolio in stages: Gradual modification of the benchmark portfolio based on expected inflows of new capital

As the fund has substantial inflows of new capital in the form of government petroleum revenue, strategy (a) would mean that the securities sold off would have to be bought back again at a later date. A strategy of this kind could therefore lead to unnecessary transaction costs for the fund. As a result, strategy (b) was considered the better option. It also opened up the possibility of spreading purchases of shares in small-cap companies over an extended period and so further reducing the expected market impact. Chart 7–3 summarises the actual transaction volume by region during the three phases in which the benchmark portfolio was modified.

Liquidity in the small-cap segment was so low that only about 50 per cent of the actual transaction volume was carried out on the same date as the change in the benchmark portfolio. The remaining trades were spread around the transition date in order to minimise indirect transaction costs, but this also increased the relative risk when the fund was overexposed to small-cap companies before – and underexposed after – the changes in the benchmark portfolio.

The transition was completed at the end of the first quarter of 2008. Chart 7–4 shows the composition of the actual benchmark portfolio on 31 March 2008. In total, transactions with a value of NOK 117 billion were performed. Norges Bank estimates the total transaction costs at NOK 1 110 million, or 0.95 per cent of the transaction value.