In this year’s Investment Competition we asked you to act as an Investment Analyst / Economist and respond to 20 demanding propositions. The competition closed for entries on 15 July. We had 163 participants who were looking to help their Portfolio Manager boss and to win bragging rights (and some fantastic prizes) across the Actuarial profession!
The ‘wisdom of actuarial crowds’
The key messages to the Portfolio Manager for the next financial year based on the majority views from participants were:
- The oil price doesn’t have much chance of rising above US$75.00/bbl
- The Japanese 10-yr bond yield is highly likely to remain negative
- The S&P ASX 200 price index is expected to remain under 6,000
- There is not much chance that 5-yr credit spreads in Australia will blow out to more than 200bp
- It is quite unlikely that Australian house prices will fall by more than 5%
- The iron ore price is expected to stay above US$40/t
- The Japanese economy is not expected to fall into near recession
- Melbourne weekend residential auction clearance rates will remain above 65% next Autumn
- US equities are expected to perform solidly, with the S&P 500 index expected to rise to above 2,200
- The AUD/USD exchange rate won’t see a lot of downside and should remain above 68.0 cents.
- China is unlikely to see a manufacturing recession with the PMI holding above 47.0
- Volatility on the Australian equity market is likely to remain modest at under 20.0
- The RBA is unlikely to cut rates more than once in the financial year
- The Fed is unlikely to raise rates more than once in the financial year
- CBA is expected to be a better performer than BHP
- The US unemployment rate is expected to remain below 5.0%
- The gold price is expected to remain below US$1,500
- The Australian 10-yr government bond yield should remain below 2.0%
- It is harder to call where Australian inflation-linked bonds will trade, but there may be modest downside from 0.90%
- We have no idea if Greek bond yields are going up or going down!
Scaling factors for each proposition
Each proposition is scaled depending on the responses from participants in the competition, somewhat like a parimutuel betting pool. The scaling factor used for each proposition is: the sum of all ‘yes’ and ‘no’ answers divided by the number of correct answers. Those participants with correct answers for a proposition will be given one point multiplied by that proposition’s scale factor. Incorrect answers will be given a score of zero points. Any ‘unsure’ answers do not affect the scale factor and will be allocated one point for that proposition.
Based on the 163 responses, the following table shows the scaling factors for each proposition:
The correlations and other details
The survey had 163 participants, including an encouraging 19 participants from outside Australia. We had two pairs of identical responses. Interestingly each of those two pairs had 17 ‘No’ responses. One participant, probably an ‘index-tracker’ type, had 20 ‘unsure’ responses. They are now guaranteed to get a ‘market-index’ 20 points. Generally, however, participants were pretty sure about their views, with the average number of ‘unsure’ responses of 1.3, with a median of 0.
Participants were most ‘unsure’ about if the S&P ASX 200 VIX index would trade above or below 20.0, with 28 participants responding as ‘unsure’. The next proposition with the most ‘unsure’ responses was if the Aussie iTraxx credit default index would be trading above or below 200. These are both relatively new contracts to the Australian financial market so there may be some element of lack of familiarity. Participants were least ‘unsure’ to the proposition that the oil price would trade above US$75/bbl with only three participants responding ‘unsure’ to that proposition.
The correlations between responses were strongest for the interest rate propositions. There was a 0.42 correlation between the responses for the propositions relating to the Australian 10-yr government bond yield and the inflation linked bond yield; and a similarly strong 0.42 correlation between the responses for the propositions relating to the Australian 10-yr government bond yield and the RBA cash rate.
Interestingly the strongest correlations to the responses to the proposition for a sharp rise in Australian credit spreads were a 0.31 correlation to the proposition of a fall in the iron ore price; a 0.29 correlation to the proposition for a rise in the US unemployment rate; a 0.28 correlation to the proposition of a fall in Australian house prices; a 0.21 correlation to the proposition of a rise in the gold price above US$1500; and -0.24 correlation to the proposition that Japanese 10-yr bond yields will remain negative.
The proposition for Greek 10-yr bond yields to trade above 8.0% was most strongly correlated at 0.28 with the proposition that Australian inflation-linked bonds would trade below 0.90%. There was also a 0.20 correlation with responses to the proposition that Melbourne auction clearance rates would fall below 65%. It’s often said Melbourne has the largest Hellenic population outside of Greece!
The responses to the proposition of a sharp rise in the gold price were correlated at 0.22 with the proposition of a steep rise in the oil price. The responses to the gold price rise proposition had a 0.22 correlation with the responses to the proposition that Japanese GDP would be below 0.5%; and a 0.21 correlation with the proposition for a pick-up in Aussie credit spreads to above 200bp.
The two propositions with responses that had no correlation above +/- 0.20 with the responses of any other proposition were that CBA would outperform BHP, and that the S&P ASX 200 VIX would trade above 20.0.
Bulls and Bears
For the basis of this analysis, and nothing to do with the competition, I have allocated the ‘yes’ responses to each proposition as ‘bullish’ or ‘bearish’. Some are pretty clear, such as if equity markets are to rise, others are more subjective – for example I have assessed a sharp rise in the gold price as ‘bearish’. Overall, I identified 16 of the ‘yes’ responses to propositions as ‘bearish’ and 4 of the ‘yes’ responses to propositions as ‘bullish’. The four ‘bullish’ ones were the rise in Australian equities, the rise in US equities, the rise in the oil price and the rise in the Fed target rate.
Overall the participants to the survey were assessed at moderately bullish with a median of two more ‘bullish’ responses than ‘bearish’ responses.
This scatterplot shows the distribution of bearish / bullish responses.
The competition was open for entries between 24 June and 15 July 2016. The Brexit vote was held within this period. There was not any negative correlation between ‘bullishness’ and the timing of entries. If anything there was a slight 0.09 positive correlation between ‘bullishness’ and the number of days into the competition when an entry was made.
Your Portfolio Manager boss is now well positioned to take advantage of the ‘wisdom of actuarial crowds’ and to make some insightful investment decisions. It is now just a matter of time to see which participants have ‘called it well’ and who ‘bombs out’. Your bonus is on the line. And remember your job is your bonus!
CPD: Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.