← All articles
Markets · Value investing · · 11 min read

Value investing when rates are high: lessons from 2008, 2011 and 2022

In March this year the interbank rate jumped from 3.7% to 9.3% in two weeks. The State Bank did not raise policy rates by a single basis point. The market raised them instead. And in July, margin accounts started blowing up.

I am not writing this to predict what the State Bank will do in Q4. I do not know. The people I read, Graham, Buffett, Schloss, Templeton, did not know either, and the interesting part is that they did not need to.

What they did differently was not forecasting better. It was changing the question.

Buffett said this plainly back in 1977

In May 1977, Fortune published Buffett's essay "How Inflation Swindles the Equity Investor". American inflation was above 6% and heading into double digits.

His argument is uncomfortably simple.

He observed that the average ROE of American business had been stuck around 12% for decades, regardless of whether inflation was high or low. Which means a stock is, in essence, like a perpetual bond paying a 12% coupon. Not 12% adjusted for inflation. Just 12%, full stop.

So what happens when a government bond pays 10%, then 14%?

Your 12% coupon bond does not suddenly pay 18%. It still pays 12%. What changes is the price people will pay for it. And you must still pay tax on the nominal interest, the interest inflation has already eaten.

Buffett called the sum of inflation plus tax rate the investor's "misery index". When that index exceeds a business's return on capital, you are getting poorer while your brokerage account is green.

The most important conclusion he made elsewhere: interest rates act on asset prices like gravity on matter. The higher the rate, the stronger the downward pull, on every asset without exception.

This is where most of us go wrong. We go looking for "inflation-proof stocks". Buffett recalculated the threshold.

Graham had a formula, and it is more brutal than you think

Graham lived through the 1970s with an arithmetic rule: only buy when a stock's earnings yield (E/P) is at least double the AAA bond yield.

Try applying that to Vietnam today.

If you treat the 12-month deposit rate as your real opportunity cost, and for a Vietnamese individual investor it is, not the government bond yield, then in September 2026 that number is:

Take 7.5% for simplicity. Graham's rule demands a minimum E/P of 15%, meaning a maximum P/E of 6.7.

The VN-Index P/E at the end of August 2026 was 12.46.

You can read that two ways, and both are correct.

One: 12.46 is considerably below the ten-year average of 15.32. The market is cheaper than its historical average.

Two: against current deposit rates, the index is not cheap at all. The VN-Index earnings yield is 8%. A Sacombank deposit is 7.5%. You are receiving an extra 0.5 percentage points in exchange for all the risk of the Vietnamese equity market.

That is not a margin of safety. That is zero reward for taking risk.

But, and this is where Graham's rule is most useful, it does not say "buy nothing". It says: the average is not good enough, so if you buy, you must be able to explain why this one differs from the average. The threshold has just been raised. Anyone who cannot clear it stays out.

Vietnam has been here three times already

What calms me is not American theory. It is our own three most recent episodes.

2008. CPI peaked around 28% in August. The base rate was 14%, deposits 18–19%. The VN-Index went from 1,170 to 235, down 80%. The bottom came about six months after rates peaked.

2011. Average CPI 18.6%, peak 23%. A 14% deposit cap, lending at 20–25%. The bottom was 336 points in January 2012. Anyone buying good businesses in 2012–2013 had a very comfortable five years.

2022. This is the most instructive, because CPI was only about 3–4%. Almost no inflation. But the State Bank raised 200 basis points in five weeks, deposits went to 9–12%, and the VN-Index fell from 1,528 to 873, down 43%. Corporate bonds froze. Highly leveraged property names lost 80–90%.

Three times, three different causes, one identical script.

And three lessons American history cannot teach a Vietnamese investor:

One, our cycles are short and violent. Not a decade like the Volcker era. Six to twelve months. But the drawdown inside that window is deeper, because retail margin dominates this market.

Two, every Vietnamese rate shock is really a property credit event wearing a monetary coat. In 2022 there was almost no inflation and the market still lost 43%. The casualty list is the same each time: leveraged developers, banks with concentrated developer exposure, and people who bought bonds without reading them.

Three, the market bottoms before deposit rates peak. In November 2022 the VN-Index bottomed. Deposit rates kept rising for two more months. Anyone waiting for the State Bank to cut before buying bought 30% higher.

The signal is not a policy decision. The signal is forced sellers.

2026: the squeeze is not coming from the State Bank

The current picture, in numbers:

August CPI up 4.89% year on year, 4.45% for the eight-month average. Core inflation in August 4.55%, eight-month average 4.24%. Transport up 8.02%, petrol alone up 15.8%. Housing and construction materials up 6.60%.

The gold price index for the eight-month average is up 46.98% year on year. That is not a number about gold. That is a number about confidence in the dong.

The refinancing rate: 4.5%, unchanged since 2024.

But market rates have risen on their own. Interbank jumped from 3.7% to 9.3% in the second half of March. Promotional mortgage rates went from 5.5–6% in 2025 to 10–11% early this year, floating at 12–14% after the promotion. The free-market USD touched 28,000 dong in late March.

The root cause is a classic gap: credit grew nearly 18%, the fastest in five years, while deposits grew only about 14%. Many banks' LDR sits against the 85% cap. Around 80% of deposits are short term, while nearly half of lending is medium to long term. Household money has flowed into gold, dollars and property.

Meanwhile Q2 GDP grew 8.4%. Exports, FDI and retail are all strong.

This is exactly the environment Buffett described in 1977: the real economy is doing well, but the cost of capital has returned, and leverage is being punished. July was the reminder: a sharp fall and margin call pressure. August recovered 5.55% to the 1,800–1,830 area, but the gains were concentrated in large caps and have not broadened.

One analyst said a line worth recording: nominal growth is now around 13–14%, roughly equal to the cost of capital. The net return left for investors, at the index level, is close to zero.

Bad news for anyone buying the index. Good news for anyone picking businesses.

What they did: six actions, not six predictions

They buy from forced sellers, not when rates fall

In November 1974, with the Dow at 600 and inflation at 12%, Buffett told Forbes: "I feel like an oversexed guy in a harem." He bought the Washington Post for about $10.6 million when the whole company was valued at $80 million, while he estimated real value at $400 million.

He was still wrong in the short run. That investment fell another 20–30%.

In August 1979, with long bonds near 10% and everyone hoarding cash, he wrote: "You pay a very high price in the stock market for a cheery consensus." The market bottomed three years later, in August 1982.

Three years. He accepted looking wrong for three years.

They accept looking wrong, but do not use leverage

Munger's fund: down 31.9% in 1973, down 31.5% in 1974. Two years in a row. Then up 73% in 1975.

The lesson he drew was not "I was wrong about value". He was right about value. The lesson was about leverage and concentration: he was right and still nearly carried out.

Contrast Walter Schloss: through the 1970s he simply bought cheap stocks on balance sheet metrics, no borrowing, no macro forecasting, and compounded at about 20% a year through the decade.

In a market where every crash is caused by margin, this is not moral advice. It is survival technique.

They change what they buy: pricing power instead of capital intensity

In the 1981 shareholder letter Buffett set out two characteristics of businesses that survive inflation:

  1. The ability to raise prices without losing volume or share
  2. The ability to grow revenue substantially with very little new capital

See's Candies, bought in 1972 at the start of the inflationary decade, is the template. It raised prices every year, needed almost no capex, and became the cash machine funding everything else.

The losers are the opposite: capital intensive, thin margins, high leverage. Steel, airlines, price-regulated utilities, textiles. Berkshire's own textile mill was on that list. Inflation forced them to reinvest ever more nominal money just to stand still.

Applied to Vietnam, the test question is not "is this a good business?" It is: over the past 24 months, by how much have they raised prices, and did volume fall? Many familiar Vietnamese consumer brands lost that ability after 2020 without their reports saying so directly.

And for banks, the group everyone says benefits from higher rates through NIM, the sufficient condition is that the loan book survives. In Vietnam the loan book leans heavily on property.

They read the balance sheet before the income statement

The 2022 casualty list was entirely predictable from three numbers: short-term debt against cash, bonds maturing within 24 months, and receivables from related parties.

Those three numbers are sitting in every listed company's H1 2026 report right now. Anyone can read them. Very few do.

One more line for this year: borrowings in USD, with the dong under pressure.

They treat cash as a yielding position, not a failure

In 1984 Buffett bought $139 million of WPPSS bonds at roughly a 16% tax-free yield, because at that yield a bond beat most businesses he could buy. In 2008 he did the same with Goldman and GE preferred at 10% plus warrants. Through 2023–24 Berkshire held $150–190 billion of Treasury bills at around 5%.

That is not standing outside the market. That is being paid to wait.

For a Vietnamese in September 2026, a 7.5% deposit against 4.45% CPI yields about 3% real. You will not get rich from that. But it is a real position with a real yield, and it preserves your optionality.

Two common errors. First: locking everything into 24–36 months for an extra 0.3 points, then having no money on the day the margin calls come. Ladder it across 3, 6 and 12 months. Second: treating gold as a form of cash. The eight-month average gold price index is up nearly 47%, which means the crowd is already in it, and gold produces nothing.

They become the lender when liquidity is scarce

Hetty Green held cash through the boom, then in the 1907 panic lent to New York City itself at above 6% while the banks froze. The largest gain of her career came from being the only one still liquid.

This is where a Vietnamese individual investor has an advantage institutions do not. Floating mortgage rates of 12–14% are creating forced sellers in the secondary apartment market right now. The masters' principle: buy from a leveraged individual who must sell, not from a developer who can still wait.

Where I could be wrong

I owe you three counterarguments.

One: this may be a liquidity squeeze, not a new inflation regime. August CPI rose 0.47% month on month, mostly from petrol. If oil normalises, this ends within months, and "hold cash as a position" leaves you out of the recovery.

Two: Graham's double rule was calibrated on American business quality. Applied mechanically to Vietnamese deposit rates it excludes almost everything, including businesses that deserve a higher multiple. Use it as the threshold for an average business, then force yourself to write out a specific argument for each exception.

Three, and most serious: 8.4% GDP growth is real. All three previous episodes ended within a year. Buffett's and Templeton's biggest mistake in the 1970s was not what they bought. It was what they did not buy.

The risk of being too Graham here is real.

The QuietVC lesson

None of them predicted the rate peak. Not one.

What they did was four very boring things: raise the acceptance threshold, read the balance sheet first, treat cash as a yielding position, and arrange their positions so that buying early could not kill them.

Then they waited for forced sellers.

In July 2026 this market gave you a preview of what forced sellers look like. If it repeats, the only question worth preparing for is not "when will the State Bank cut".

It is: where is my list, and how much money do I have left to buy?

That list can be written today, before anything happens. That is almost the entire job.

Silent discipline. Compound value.

Hoan Do · Founder, QuietVC