Intermarket analysis and sector rotation

Reading for India · about 4 min

What this cluster is for

Everything before this cluster looked at one instrument at a time. This cluster looks at the relationships between them, and it answers a question most investors never ask: what is my portfolio actually a bet on?

The answer is almost never the list of names in your account. A portfolio of 8 holdings usually turns out to be 2 or 3 bets wearing 8 labels, and finding that out takes about 10 minutes.

This subject is more defensible than the classical methods in cluster 18, and it should be read that way. The connections between bonds, shares, commodities and currencies have real economic mechanisms behind them. Almost all of them run through 2 variables: inflation and the interest rate. That is a genuine explanation, not a chart pattern.

But 2 honest limits run through every article here.

The sign of a relationship is not fixed. Shares and bond prices moved together in the United States for roughly 30 years, then in opposite directions for roughly 20, then together again in 2022. A correlation is not a law. It is a readout of which kind of shock the market is currently reacting to.

The sector rotation clock predicts weakly. It fits history loosely and its essential input — which phase of the cycle you are in — is only identified long after the decisions that depended on it. The committee whose job is to date American recessions took roughly 15 months to announce that the 2020 recession had ended.

So this cluster keeps the measurement and drops most of the forecasting. The measurement part is free, takes minutes and is reliable. The forecasting part is what gets sold.

By the end of this cluster you will be able to:

Name the mechanism behind any market relationship you rely on, in 1 sentence, or discover that you cannot
Test a relationship you believe in by counting 12 months of moves instead of looking at a chart
Say which of 4 dominant drivers is operating now, and read the signs from that rather than from a rule
Group your own holdings by sector and see the macroeconomic position you never chose
Build 2 ratios for every holding — against the index, and against its own sector — and find the ones that look successful and are not
Read a relative rotation graph and say exactly what it adds, which is convenience and not information

The reading order

Three articles. The first supplies the mechanism, the second applies it to sectors, and the third gives you the measuring instrument.

Intermarket analysis: how markets move together — the 4 classic links, why every one of them runs through inflation and interest rates, and why the sign of each depends on which shock is dominant.
Sector rotation and the business cycle — the clock, described properly, and then the problem: the phase is a label applied afterwards. What survives is measurement of current leadership, which needs no economic call at all.
Ratio analysis and relative strength — one price divided by another. Direction is the message, level means nothing, and 2 ratios per holding separate a sector skill from a company skill.

The checklist

Three exercises, about 25 minutes in total, all on your own account.

On a relationship you believe in (1)

Write the relationship in 1 sentence before looking at any chart.
Count 12 month-end moves in both series. How many matched your relationship?
Repeat for the 12 months before that. Compare the 2 counts.
Write the mechanism in 1 sentence. If you cannot, you hold a correlation, not a relationship.

On your holdings, by sector (2)

Write each holding's current value, not its quantity and not its profit.
Label each with its sector, and total the value in each sector.
Divide by total portfolio value including cash. Sort largest first.
Add the top 2 percentages. Above 50% is a concentration you probably did not choose.
Work out what the portfolio is worth if your largest sector falls 30%.

On your holdings, by ratio (3)

For each holding compute: stock change − broad index change over 6 months.
Then compute: stock change − its own sector index change over the same 6 months.
Sort by the second number and place each holding in 1 of the 4 boxes.
Positive against the index and negative against the sector: the sector is carrying you. Write down why you hold this specific company.
Negative on both: write 1 sentence saying why you still own it. If it mentions your purchase price, it is hope.

Habits that cost nothing

Before treating 2 markets as related, name the driver: growth improving, growth deteriorating, inflation rising, or panic.
Size positions as though correlations go to 1 in a crisis, because they do.
Check that both legs of any ratio are the same kind of series — both price return or both total return, both adjusted for corporate actions.

India and the United States, equally

Each article covers both markets at the same depth, then says what the difference tells you. In this cluster those sections are unusually practical, because the frameworks were all built in the United States and 3 of them do not transfer.

India's sector map does not match the clock. There is no large, liquid utilities sector to rotate into and real estate is a small part of the index. Financial services carries a very large weight in the Nifty 50. Several Indian sector indices are dominated by 3 or 4 companies, so buying a sector here is much closer to buying a few companies than it is in the United States.

India's economy is less synchronised with the United States than commentary assumes, and its asset prices are more synchronised than its economy. Indian growth is driven substantially by domestic demand, so the real cycle can diverge. Indian share prices are set at the margin partly by foreign portfolio flows, which respond to American interest rates. You can be right about the Indian phase and wrong about the Indian market.

One flow variable moves 3 Indian markets at once. When foreign investors sell, they sell Indian shares, Indian bonds and rupees together. An Indian investor holding all 3 holds 1 exposure. That is why net foreign portfolio flow, published daily and free, is closer to a master variable in India than any single number is in the United States.

And the conclusion is purchasable in one market and not the other. An American who finds a leading sector buys the sector fund at a spread of a cent or two. An Indian who finds a leading sector usually has to pick companies inside it, which takes on risk the sector analysis never measured. That is why the 2-ratio test is not optional here.

A note on what this is not

These lessons are free and they will stay free. They are the text version of the ideas in our video course, written for anybody who cannot spend money on learning right now.

Nothing in this cluster forecasts a market. What it does is tell you what you already own, in terms of the forces that will actually move it. That is a smaller promise than the one usually made about intermarket analysis, and it is the part that holds up. The goal is the same one it has always been: fewer people in the 95%.