20 October 2011 gold ETF report: test of faith

Today my ETF Securities Physical Gold (PHAU) shares dropped 2.4%.

I'm not going to interpret the causes - some of them are discussed here (IB Times) and some more here (MarketWatch). The argument is that if gold isn't a fear barometer it's a useless investment for a while.

Where do I stand in this situation?

First the boring bit - deciphering what's happened to my PHAU shares. Then I'll assess whether I'm in enough pain to act and, lastly, whether I have enough knowledge to act sensibly.

When I bought the shares on 22 September I paid $172.734 per share (that's before adding the £11.95 I was charged to carry out the transaction.)

At the close of business today Hargreaves Lansdown would have paid me $158.03 for each of them. (The offer price was $158.22 per share.)

So my shares have dropped 8.5% in their dollar value - again that's before adding the cost of dealing.

Unfortunately that's the happiest spin my gold ETF story gets today.

When I bought the shares they cost £111.9787 each. There were 8 of them which adds up to £895.8296.

By the end of today - 20 October 2011 - my eight shares had a sterling value of £803.81 according to HL which used an exchange rate of $1.5728 to £1 to get this figure.

So in sterling, before dealing costs, my loss is £895.8296 - £803.81 and on that count I'm down £92 which equates to a 10.27% loss. That's a lot worse than the 8.5% dollar loss calculated above.

That appears to be because the pound has strengthened against the dollar (check this Google chart). It now costs me more to buy pounds with my gold ETF dollars than it did back on September 22.

At the end of 22 September £1 would have bought me $1.5358. Now, nearly a month later, my pound buys me $1.5728. That works out at a 2.4% increase in the purchasing power of my pound against the dollar. That has worked against me because my investment is in a dollar denominated asset.

If I wanted to sell now I'd have to include a £11.95 transaction cost which would come off what HL would pay me: so £803.81 - £11.95= £791.86.

When I first bought the shares I also paid £11.95, a total of £907.8 for my shares and so the difference between the entrance and exit price widens again. Now I would be exit with £791.86 when I spent £907.78 entering. A £115.92 loss or -12.7%.

So what do I do?

It feels like I picked a testing time to try out a 'leap-before-you-look" investment style in gold. But where does my faith lie? Should I sell and wait til things get better? Or should I hang on?

I'm a financial journalist of sorts (out of work at the moment!) and I've written about gold related investments for a while. This generally involved passing on various people's views about the price of gold and silver.

It's embarrassing but while writing about it, it didn't cross my mind to ask such a basic question as: "Where does the price of gold come from?"

But when I bought physical gold ETF shares it was an obvious hole in my knowledge because my investment tracks the "spot price" of gold.

Adrian Ash, head of research at the BullionVault helped me work towards an answer in this piece: Is the gold "spot price" real?

Then today I read some of this blog "Gold Chat" and was relieved to find out that it's quite common for people to take the construction of the gold price for granted. They don't realise that it's an issue until they buy the stuff.

Bron, who works at Perth Mint in Australia and writes Gold Chat said: "It was always amusing to me when clients would ring up to buy and we would quote a price and then, naturally, they would say “Well, where can I get what the spot price is?” so they could work out if our price was “fair”. The answer was, “It doesn’t exist.""

I expect there will be lots more surprisingly basic (or stupid) questions I need to answer before I get comfortable with the product I own, let alone the gyrations of the gold market.

On that front Bron is a reassuring read for people left confused and worried by the daily shifts in the gold price - look at the day trading section of the Investment Time Frames section. I feel like this has provided me a bit of space to work out what my motivations were for buying gold in the first place. I'll be looking at Gold Chat's investment time frames: part one and part two to start off with.

I don't know what I'm doing but, at the moment, owning gold has forced me to pay attention. I don't think I'd be interested if I wasn't invested enough for this to be a painful experience.

I'm not interested in diversifying until I have a better grasp of what I've already got - although I do own BlackRock Gold and General Fund too.

Is the gold "spot price" real?

I don't own gold I own the 'spot price' of gold

After taking a look at the prospectus for my ETFS Physical Gold (PHAU) shares I'm fairly sure that I'm not really the owner of gold. Yes the shares are backed by real lumps of the stuff in a vault but the chances are that I’ll never get my hands on it.

In reality the gold backing my PHAU shares is more like collateral - it’s what I have a right to if all else fails. Before that happens though, the prospectus for ETFS Physical Gold says my shares are valued at something called the “spot price” of gold (ETFS said that PHAU tracks the loco London spot market).

So what is this “spot price”?

On the face of it, the construction of the gold price appears a bit random. Adrian Ash at UK online gold and silver market BullionVault told me that the price is not standardised, which means it’s not like tracking the share price of Marks and Spencers for which there is a single price quoted through the London Stock Exchange.

So, in the case of gold, the source of the price is not set in stone. Ash says that the data feed that supplies the free online "spot" chart at BullionVault, where he is the head of research, is not likely to be the same as the data feed that supplies, say, the chart at Canadian bullion dealers Kitco and the prices may be slightly different.

This is because the companies that provide the data feeds can pick and choose the sources of the data. These will be dealers who have agreed to pass on their live buying and selling prices. However Ash points out that any difference in price will be minimal because a dealer who deviates far from the global price will either have everyone knocking on their door… or no one.

In general though, the prices that make up the core of the global spot price come from the big banks listed as the “market makers” for the London bullion Market Association – centre of the world's wholesale physical trade.

These include the likes of JPMorgan, HSBC and to be an official ‘market maker’ an institution promises to offer prices at which they will buy and sell gold at all times during the hours of London trading.

But these banks are global so when trading shuts in London it shifts over to New York.

Wherever the trading is going on, the prices offered are being fed to the likes of Bloomberg and Reuters and other data providers who then compile gold “spot prices” for their clients. To do this they take the mid points between the buying and selling prices offered by each market maker and find the average.

Ash said: “So take note – any "spot" price data you see will fail to show any widening of the spread between buying and selling prices during strong volatility.” So investors should look carefully at the difference between the price they are being offered and the gold spot price.

There are further complications to the foundations of the “spot price” and the extent to which it describes the price of physical gold. To start with, the London market makers are not the only sources of data or the only influence on the gold price.

But during London trading hours they are the biggest players and the deals being done in London are for metal and are supposed to be completed within two days. But that’s not the case in other markets that influence the gold price. Ash said that when trading shifts to the USA the biggest gold market open for business becomes the New York futures market – where promises to deliver gold at a specific date are traded, but the vast bulk of contracts are in fact settled for cash, not metal.

Influences like this are somehow factored into the data feeds for “spot gold” prices supplied to traders.

In contrast to the apparently laissez faire price setting – where there can be minor discrepancies – there is no flexibility about the gold that is bought and sold in the loco London spot market.

This is entirely standardised with approved refiners, vaults and traders. If the gold is in the system it is considered good for delivery and if it leaves the system it isn’t allowed back in without severe checks.

Conclusions: Supply and demand, investor fear, interest rates and the value of the dollar are the engines that drive the gold price. But the gold price itself is the speedometer and it seems sensible to check that it's actually measuring what you want it to measure.

The fact that it doesn’t include spreads - the difference between the price at which gold is bought by market makers and the price at which they sell it – essentially means that gold is will not change hands at this price.

It may be useful to find out more about when and why spreads widen and whether the changes are large enough to let them influence an investment decision.

Also, does anyone monitor the differences between gold price data feeds, particularly in times of crisis? Are there any technical dangers in terms of the gold price?