Professional investors have successfully relied on negative correlation between stock and bond prices for more than two decades. When times are bad for stocks, bonds go up, and vice versa. These two instruments cancel each other out, and as a result, hedge fund clients are pleased with stable returns.

For example, the average correlation over the past two decades has been negative 0.3, which means that falling stocks have often been accompanied by a rally in bonds.

However, everything has changed.

For over a year now, analysts have been sounding the alarm that the era of negative correlation appears to be coming to an end. It is being replaced by an era of positive correlation, in which bonds and stocks tend to move together in the wake of the general sentiment of traders (and increasingly, this starts with retail players). At the same time, the total volatility of any investment portfolio that is based on these two positions is increasing, not decreasing. Bonds have never been so useless as a safety rope against the stock market (since the 1999 crisis - ed.)

What is the reason? Fear of inflation.

The role of Treasury bonds as a shock absorber was undermined with the introduction of the Fed's zero interest rate plan as fear of inflation united both equity investors and bond investors. If it persists, it would represent a dramatic change, as strategies such as risk parity and 60/40 will no longer be reliable.

In May, as inflation fears peaked, Charlie McElligott, a miscellaneous asset strategist at Nomura Securities, wrote: "...now, due to the pandemic response, that old dynamic simply no longer applies. Inflation is a volatility catalyst. Only in the case of an extreme inflation overshoot would the Fed's hands be tied, forcing it to raise rates more quickly and crash both bonds and stocks."

But the market has been living on hope for the last two years.

The yield on bonds has been falling steadily over the past week. What does this mean?

First of all, in this way, investors signal to the market that they have decided to trust the Fed's promises and are optimistic. Swap traders expect the next cycle of central bank price increases to end at rates well below the so-called final rate - a level that will be neutral to economic growth. Forward data showing where monthly rates will stand in five years has fallen far from this year's high in late March and far from the midpoint representing the Fed's long-term forecasts.

Taken together, it suggests that investors have decided, at least temporarily, that inflationary risk is temporary (but the pace of economic recovery will also be sluggish).

The successful rally in Treasury bonds instantly revived the familiar stock-bond relationship that underpins the main diversification strategies used by investors around the world.

The 20-day correlation between S&P 500 futures and Treasuries turned negative for the first time since February, according to data compiled by economists Bloomberg. While long-term readings remain positive, they still signal a reversal of the recent trend.

This is largely due to the not too rosy US employment report, according to which employers are experiencing great difficulties in attracting labor back to vacancies.

The market is gradually realizing that the recovery will not be easy, and the trading on reflation is not living up to expectations.

As a result, the yield on 10-year US bonds closed 8 basis points lower, falling more than 17 basis points since the beginning of last week.

However, the correlation has been declining for now, and the Treasury's latest rally did not coincide with a sell-off in stocks. This is good short-term news for portfolios using both. The $1.3 billion RPAR Risk Parity Exchange Traded Fund (ticker RPAR) set a new record on Tuesday. DFA's Global Allocation 60/40 portfolio is also close to all-time highs.

You need to understand that in the long term, a positive correlation between bonds and stocks is needed. After all, these are competing investments, since each of them forms its own stream of constant income: dividends on shares, coupon payments on bonds.

If stocks get very expensive, investors flop to the bond market, forcing them to rally until they become too expensive. Then, investors prefer to return to the stock market.

This free market rebalancing keeps the value of both assets in perspective.

Therefore, when we talk about negative correlation, we mean first of all short periods (from one month to three).

Recently, this scheme has been broken.

According to a spokesman for Bridgewater Associates, the world's largest hedge fund based in Westport, Connecticut, stocks and bonds do not exist in a vacuum - they are directly related to the market situation: "Naturally, there will be times when they are negatively correlated, and naturally there will be times when they are positively correlated and they come from the most basic environment."

According to Gardner, inflation has been the most important factor in the markets for decades - both when it rose in the 1960s and 1970s, and when it fell in 1980s and 1990s. Indeed, it is difficult to imagine a more powerful fear that makes ruined people sometimes even contemplate suicide.

Inflation affects stocks and bonds in the same way, but it is worse for fixed-payment bonds than for stocks. Therefore, the correlation was positive over this extended period.

Since 2009, inflation has been so low and steady that it has not affected the markets. So, investors began to pay more attention to the prospects for economic growth. Strong rallies are good for stocks, but do nothing for bonds. This, according to Gardner, is the main reason that stocks and bonds have diverged in different directions.

PGIM Inc., the main asset management business of the insurance company Prudential Financial Inc., manages $1.5 trillion. In May, they were among those voicing concerns about the robust positive correlation.

It can be understood by looking at the numbers.

Let's say a portfolio is made up of 60% stocks and 40% bonds and has a stock-to-bond ratio of -0.3, which is roughly the average over the past 20 years. Volatility is around 7%.

Now, suppose the correlation tends to zero - not yet positive, but no longer negative.

To keep volatility from rising, the portfolio manager will have to cut the equity allocation to around 52%, which will reduce the portfolio's return.

If the correlation between stocks and bonds reaches a positive value of 0.3, then to prevent the increase in volatility, it will be necessary to reduce the distribution of shares to 40%, which will further reduce the yield.

PGIM's list of factors that influence correlations is longer than Bridgewater's, but consistent with it in key points. Vice President Junying Shen and Managing Director Noah Weisberger said the correlation between stocks and bonds tends to be negative when there is strong fiscal policy, independent and rule-based monetary policy, and when consumption levels change in the economy (sustainable growth - ed.)

According to them, the correlation is likely to be positive with unsustainable fiscal policy, discretionary monetary policy, coordinated monetary policy, and a dominance of supply over demand.

It's hard to disagree with them. The surplus of the "cheap dollar" has led to the fact that investors are happy to buy up even junk assets, both in the sector of stocks and in the sector of corporate bonds. The quality of securities, as expressed by ratings, has never been so low.

Nevertheless, the balance between stocks and bonds is worth keeping, even if they are not hedging each other.

For example, there is an interesting capital asset pricing model developed by William Sharpe in the 1960s, which states that everyone should have the same portfolio of all available assets and adjust their risk based on the amount borrowed. Hedge fund managers, of course, disagree with this overt socialism, which means profit averaging. Besides, if no one really goes bankrupt, then no one really wins, right? For the market, suspension of cash flows means stagnation, inflation and death.

Let's not forget that the real base interest rate is actually negative. This means that the holders of the benchmark 10-year bonds will not receive income from their investments for another 10 years. Taking into account the extremely slow pace of economic recovery, we may again reach a positive correlation in the fall, if it turns out that the surge in business activity usual for this season will be more sluggish. And if such a correlation lasts until spring, which is quite possible given the still strong quarantine measures, then this could kill investor confidence in bonds in principle.