“This keeps portfolio volatility  constant and we can measure how much  return for risk we are taking,” Iverson  says.  This discipline is similar to the  “total portfolio” approach used by the  $153-billion Canada Pension Plan  Investment Board (CPPIB). It stipulates  that any move away from its policy  portfolio (which acts like NZ super’s  reference book) to its real portfolio must  be funded by a corresponding shift in  assets within the policy book.  NZ Super and the CPPIB are also  similar in risk profile: contributions to  the Canadian fund are expected to exceed  annual paid benefits until 2021 and its  investments are planned to amortise over  a 75-year period.  NZ Super’s governance principles  allow the investment team to respond  autonomously to opportunities.  However, more complex investment  decisions require board oversight and  most of the asset allocation ranges it  operates within are a legacy from the days  when the fund adhered to strategic asset  class weights.

“With the reference portfolio  construction, investment limits  become less relevant,” Iverson says.  “Portfolio expectations are more  important.”  These expectations are very  fluid – they are largely dictated by the  opportunities presented by markets. While the fund does not want private  equity investments per se, it asks if  the risk exposures currently offered  by the strategies will help it meet  return objectives.  In addition NZ Super doesn’t  rebalance its portfolios, so the  difference between the reference  and real books is constantly shifting  because of market movements.  Active investor  The fund actively invests in  listed and private market assets such  as infrastructure, private equity,  timberland, property and farmland.  In 2008 four guardians, or  trustees, of NZ Super – Aaron Drew,  Richard Frogley, Tore Hayward and  Rishab Sethi – published a paper  entitled “Strategic tilting around the  SAA benchmark”. This technique  has become another way in which the  fund invests actively.

According to the paper, “in  the presence of uncertainty over  the return predictability process,  strategic tilting tends to perform  at least as well as the traditional  approach of rebalancing asset classes  to their weights in the strategic asset  allocation”.  But there is a catch. Tilting too  early can result in poor short-term  performance before longer-term  gains are enjoyed. But the New  Zealand Government will only begin  drawing on the fund in 18 years so it  can invest with a genuine long-term  horizon.  “We are still in the early days of  strategic tilting,” Orr says, adding that  it can be difficult when your client  is the Government, whose actions  are routinely reported in the press.  “You are taking risk on when it feels most  uncomfortable, but that is exactly what  you should be doing.”  The fund’s total economic exposure  to asset classes at the end of the most  recent financial year diverged from the  reference portfolio. Its allocation to global  equities was 60.6 per cent against the  70 per cent in the reference portfolio. Its  exposure to domestic equities was 5.2 per  cent; fixed income 11.3 per cent; listed  property 5.5 per cent; and private equity,  infrastructure, timber and other private  markets amounted to 20.1 per cent.