The US consumer price index rose significantly to 3.1% in April from April 2010 and by 0.4% from its March level. The increase was significantly lower than the consensus, but the figure was in line with my expectation of 3.2%. I am of the opinion that the US CPI inflation rate for May will come in at approximately 3.4% with the month-on-month increase falling to 0.1% from April's 0.4%. That compares to 0.5% in both February and March. I believe that May's figure on the year-ago figure would herald the peak in the current cycle.

Let me explain. Most forecasters take their lead from the ISM manufacturing and non-manufacturing PMI price indices. Generally speaking the trends are reasonably in line, but significant diversions do appear.

Sources: ISM; I-Net Bridge; Plexus Asset Management.

My analysis indicates that changes in the CPI inflation rate on a year-ago basis and especially CPI ex shelter inflation are explained by changes in the oil price compared to a year ago. This is particularly evident since the end of 2006 where more than 94% of the direction of the CPI ex shelter can be explained by the year-on-year absolute change in the price of crude oil.

Sources: I-Net Bridge; Plexus Asset Management.

Shelter's weight of approximately 32.3% in the CPI and the one-month lag between the change in the oil price and CPI ex shelter inflation enables me to make a reasonably accurate forecast. The change in the oil price is a known factor, while the only unknown factor is the CPI shelter number but with a more stable trend, it is fairly reasonable to assume what CPI shelter inflation rate can be expected.

Sources: Bureau of Labor; Plexus Asset Management.

The year-on-year change in the Light Louisiana Sweet crude price in April was $38.43 per barrel. The historical relationship between the change in the Light Louisiana Sweet crude price and the CPI ex shelter inflation rate points to a year-on-year CPI ex shelter inflation rate of 4.55% in May, given the said lag. That makes up 67.7% of the overall CPI inflation rate (overall CPI minus the 32.3% weight of the shelter CPI) and will therefore be 3.08%. If I assume a year-on-year change of 1% in the shelter CPI, May's total CPI will add up as follows:

(67.7% of 4.55%) plus (32.3% of 1%) = 3.08% plus 0.323% = 3.40%.

But where is CPI inflation heading?

Without taking a stab at where the oil price is heading, I looked at three scenarios where the price per barrel of Louisiana Sweet crude was kept constant at $115 (current), $130 and $100 respectively for the next 12 months. The monthly oil price was then compared to the price a year ago and depicted against the CPI ex shelter inflation rate lagged by one month.

Sources: Bureau of Labor; I-Net; Plexus Asset Management.

By applying the historical regression equation, the trend of future CPI ex shelter year-on-year inflation is as follows:

Sources: Bureau of Labor; I-Net; Plexus Asset Management.

Assuming that the year-on-year inflation rate for shelter remains steady at 1.0%, the outlook for the overall CPI inflation rate is as follows:

Sources: Bureau of Labor; I-Net; Plexus Asset Management.

From the above it is evident that the US CPI inflation rate is likely to peak at 3.41% in May if the oil price maintains its current level or weakens. Even if the oil price spikes to $130 per barrel and maintains that level, the inflation rate will top out in June this year. There may be a brief further spike in September if one takes into account the brief drop in the oil price in August 2010.

Is now the time to reconsider one's precious metals holdings?

Only about 50% of the year-on-year change in the gold price in US dollar can be explained by the change in the oil price over the same period, but it is clear to me that they share the same trends. In the following graph I have depicted the two series together with the year-on-year change in the oil price, given the scenarios of the oil price remaining constant at $115, $130 and $100 per barrel respectively.

Sources: I-Net Bridge; Plexus Asset Management.

I think it can reasonably be expected that the year-on-year change in the gold price will be likely to follow the trend of the year-on-year change in the oil price.

To determine the year-on-year change in the gold price given the oil price scenarios, I used the regression equation of the said relationship between gold and oil.

Sources: I-Net Bridge; Plexus Asset Management.

But what does this hold for the future of the gold price, you may ask. The future year-on-year changes in the monthly gold prices were added to the respective prices a year ago to determine the future monthly gold prices.

Sources: I-Net Bridge; Plexus Asset Management.

The gold prices listed above compare to the latest close of gold spot (New York) of $1,496. Yes, I accept that this analysis could be controversial, but at least it gives me a good feel for where gold is heading. It is difficult to see gold continuing its strong momentum, especially in light of the imminent peaking of the US CPI inflation rate. The oil price has to continue rising strongly to ensure a substantial further rise in the gold price, unless another global financial disaster strikes.

What is also worrying is that, insofar as the relationship between gold and oil as described above is concerned, the beta of the regression equation stayed relatively constant over various periods since 1998, but the alpha has increased by approximately $100 since 2007. Will gold lose this premium as well when the US CPI inflation rate starts the down leg of the current cycle?

Is this the logic George Soros followed when he decided to lighten his gold holdings?

Did you enjoy this post? If so, click here to subscribe to updates to Investment Postcards from Cape Town by e-mail.