Gold draws emotional responses from investors, whether they have a long-term or short-term time frame. While some believe that gold is one of the best ways to defend your portfolio against the whims of government monetary policy and floating exchange rates, others point to the unproductive nature of the yellow metal as reason to avoid including it in an asset allocation model, in favor of income-producing assets like stocks.

Regardless of which side of the fence they're on, though, many investors forget that at its root, gold is a commodity and therefore has supply and demand fundamentals that help drive its price. Yesterday, a new report from the World Gold Council shed some light on the trends the affected the gold market in 2012, and what the report says raises some interesting questions about where gold could go in the future.

Who's buying gold?
In general, there are four main drivers of gold demand. Jewelry and private investment represent the two major components of demand, with uses in the technology industry representing a much smaller element.

For the most part, demand from these sources tends to be relatively stable from year to year. For instance, in the jewelry industry, gold volume fell 3% during 2012 from the previous year, although higher prices meant that the value of that demand rose by 3% to a record $102 billion. Similarly, private investment volume fell 10%, despite a better than 50% jump in ETF investing from SPDR Gold (GLD -0.25%), iShares Gold (IAU -0.29%), and other exchange-traded bullion vehicles. Gold volume for technology purposes fell 5% but rose very slightly in value terms to set a new record as well, with electronics using the lion's share of the metal.

But the fourth driver of demand is also the most volatile: central bank buying. Around the world, central banks added a net 535 metric tons of gold to their coffers last year, representing a 17% increase over 2011's net purchases. Although it wasn't enough to lead to volume growth in 2012, central bank buying has been a major influence in the gold market over the past three years.

Where's the gold coming from?
On the other side of the equation, gold supplies fell slightly, as recycling activity fell and producers on the whole hedged less of their production. Moreover, the mine production picture was mixed, with labor problems in South Africa leading to declining supply while other areas, including Russia and China, boosted gold production levels.

The report's figures are consistent with trends among gold mining companies. Around the world, many miners have deferred bringing new production online due to high costs and labor issues. Just in the past year, Barrick Gold (GOLD 1.64%) and Kinross Gold (KGC 1.39%) have both made decisions to hold off on multibillion-dollar mine development projects, and producers across the industry have had to deal with cost pressures and worker unrest.

A good sign, or a sign of a top?
With strong fundamentals, the gold market seems healthy from a supply and demand perspective. But one problem throws cold water on the bullish argument: Central banks have had notoriously bad timing in their purchases and sales of gold, and so it's tempting to see their recent buying as a contrary indicator.

Look, for instance, at the report. From 2003 to 2009, as gold prices were steadily and significantly rising, central banks sold a net total of 2,880 metric tons of gold, representing nearly a year's worth of total demand during that time frame. By 2010, when governments became net buyers, gold had already risen to the $1,100 to $1,400 per ounce range.

Put another way, when you look at the cash flow at central banks based on gold value, you'll find that over the past three years, central banks have spent $6.3 billion more to buy gold than they got from selling gold during the seven prior years. Moreover, even though they spent more money, they bought back only 1,069 metric tons of gold since 2010 -- less than 40% of the amount they had sold previously. That doesn't show good timing, and contrarian gold investors have to worry that central bank interest is as much a sign of a reversal as it was when central banks sold near gold's lows in the 1990s.

What to do
Having exposure to gold makes sense in a low-interest-rate environment in which the costs of carrying nonproductive bullion are relatively small. With mining stocks having done badly over the past year, however, the potential for both share-price appreciation and possible dividend income from owning high-quality gold miners individually or through exchange-traded mining-company vehicle ASA Gold and Precious Metals (ASA 2.00%) and similar funds looks more compelling than bullion -- regardless of whether central bank buying turns out to be an ill omen going forward.