Data for June has been released, showing that the central bank increased its gold holdings by 14.93 tonnes, accelerating from May's 9.95 tonnes and marking one of the highest monthly purchase levels in recent months. Since November 2024, the central bank has maintained a buying pace for 20 consecutive months. Whether gold prices were at highs or in a correction phase, the overall buying rhythm has not experienced any significant interruption.

If we examine these 20 months more closely, the buying pace can be broadly divided into three phases. The first phase began with a restart in November 2024, followed by a rapid six-month buying spree. The second phase, from March 2025 to February 2026, saw a slowdown in purchases. The third phase started with a sudden acceleration in March 2026, culminating in a new high this June. I suspect the same team is making these decisions, as the buying rhythm for gold mirrors that of the state team's stock purchases: being cautious when prices rise most sharply, and continuously buying the dip when prices fall. The only difference is that with gold, the decision is only about buying more or less, with no selling; stocks are different, and if they are deemed overvalued, they can be completely liquidated.

Although the central bank has significantly increased its gold purchases in recent years, its proportion relative to our foreign exchange reserves is not high, currently standing at only 10%. The global average for central banks is 28%, with countries like France and Germany achieving a ratio of up to 70%. With the decoupling between China and the US, I estimate that the central bank will continue to aggressively buy gold for some time, at least until the gold ratio is raised above the global average level.

Gold is a difficult asset to value because it generates no earnings, making metrics like PE, PB, and DCF useless. The only somewhat useful indicator is the cost of gold mining. However, this can only determine gold's minimum value, not its fair price or maximum price. Therefore, for gold, one needs to focus more on capital flows, sentiment, and some grand narrative logic. For instance, I wrote articles recommending gold for an entire year last year, but why have I written less in the first half of this year? Because at the beginning of the year, gold was extremely hot, changing price by the day and rising dizzyingly. Everyone was rushing to buy gold at that time. I remember having to queue for at least 20 minutes at a Laopu store, which is completely different from the current situation where there are no queues. Many articles were analyzing why gold could rise, urging people to buy, and a lot of capital flowed into the capital markets, pushing gold prices higher. Typically, experienced researchers know a short-term overheating is inevitable in such situations.

I have a view: an asset that goes viral will inevitably peak rapidly in the short term and then crash. For example, the recent "Korean Girl" craze, which was followed by a rapid peak in Korean stocks and a significant pullback now. So, for such viral assets, I generally have two suggestions: either don't touch them at all, which is the safest option; or buy immediately and sell within a week (preferably within three days), betting on a final speculative rally. When gold went viral, I sensed that after the herd funds dissipated, there would be a decline. So, even if gold's long-term logic is sound, the short term should be avoided. But now the situation is different. Gold has corrected by about 30%, and the correction has lasted for four months. In terms of magnitude and time, short-term funds that needed to exit have already done so. It is only when the short-term is safe that discussing the long-term logic becomes valuable, so it's time to talk about gold's long-term value again.

Historically, gold has experienced three major bull markets, each with a different logic. The first bull market was in the 1970s, when the Bretton Woods system collapsed and gold was decoupled from the US dollar. This coincided with several oil crises. In this super bull market, gold rose from $35/oz to $850/oz, a total increase of about 23 times. The logic for this rise was twofold: first, the decoupling from the dollar, which had artificially suppressed the gold price at $35/oz, allowing it to rise to market levels; second, the occurrence of crises, during which gold appreciated most rapidly, making crises a historical opportunity for gold.

The second bull market was in the 2000s. After the financial crisis, the US implemented a policy called "Quantitative Easing," which essentially meant printing money indefinitely until the economy improved. The main logic for gold's rise was inflation. Although the dollar was decoupled from gold, these two remain the largest international reserve currencies. When the dollar is over-issued, gold naturally appreciates to match the global circulation of dollars.

The third bull market is from 2022 to the present. The trigger for this bull market was the US weaponizing the dollar to sanction Russia, which directly led to a decline in dollar credit. The main logic for gold's rise is its role as a substitute for the dollar. This is why we see central banks around the world accelerating their gold purchases, with the People's Bank of China being a clear example, as gold's share in foreign exchange reserves continues to increase. Of course, while I have identified the main logic for each phase, other logics also exist concurrently. For instance, in the current bull market, the primary logic is replacing the dollar, but the logic of dollar over-issuance and inflation also persists.

According to a report from the World Gold Council (WGC), major global central banks have no plans to strategically sell gold in the next five years. The general direction is to continue increasing gold holdings (excluding tactical sales for fundraising purposes), with the goal of continuously raising the proportion of gold in foreign exchange reserves to hedge against the risk of declining dollar credit. The main buyers are emerging countries, which historically have low gold ratios in their reserves. The current global weighted average central bank gold reserve ratio is 28%. In comparison, China's is only 10% and India's is 17%, leaving significant room for further increases. This may also explain why the central bank will sell stocks if they are overvalued, but will only buy less gold if it is overvalued, never selling. Gold is a strategic asset, crucial for national financial security. In such a situation, it would be very difficult for a leader to justify selling gold to higher authorities.

Finally, let's discuss why it is recommended now. Besides the long-term logic mentioned above, there is also a short-term catalyst: the renewed conflict between the US and Iran. Generally, at the onset of a crisis (like war), a liquidity crisis causes all global assets, including gold, to fall as investors sell to raise US dollars. However, as time passes, some assets (like stocks) may continue to decline, while gold, after a brief drop, tends to rise again. The essence is a shift from a liquidity crisis logic to a safe-haven logic. So, this new conflict presents a short-term opportunity. Gold has fallen by 1.5% today, but this should not last long. Once gold stabilizes in a few days, it could present a buying opportunity. For those who are less experienced, dollar-cost averaging is a good strategy, buying more as prices fall. Whether from a short-term or long-term perspective, gold appears to present an opportunity going forward.