USD/COP · SPECIAL, PART 2 OF 2
Elliott Wave Street · Part two. If you missed part one, start there.
Yesterday I left a number hanging on purpose: for every 100 pesos of growth in foreign holdings of Colombian government bonds over the past year, 86 came from the same pocket.
Today I’ll tell you whose pocket that is — and why the number that actually matters is not how much it bought, but what everyone else was doing while it bought.
A note on figures: Colombian amounts are enormous in local currency, so I have converted them to dollars at approximate rates. The peso moved sharply over this period, so treat the conversions as indicative rather than precise.
December, and an unusual sale
In December 2025, Colombia’s Finance Ministry placed 23 trillion pesos of TES bonds — roughly $6 billion.
There was no open auction. The bonds went to a single bidder and terms were agreed directly. The rate was reportedly 13.15%. The buyer emerged afterward: Pimco, the American asset manager founded in 1971 by Bill Gross, now a subsidiary of German insurer Allianz, with more than $2 trillion under management.
The Ministry defended the placement, saying Pimco’s offer came in below the quotes it had received and that a direct sale avoided moving the market. The absence of an auction drew criticism. That argument is still running and it isn’t the one I care about today.
There is another detail from that placement that does interest me, and it reads differently now. Colombia’s head of Public Credit at the time stated that Pimco’s managers had committed to holding their position — a buy-and-hold strategy — and that he would monitor the portfolio to verify they kept their word.
That commitment never made it into the contract. It was verbal.
What I care about is what happened next.
Pimco didn’t stop there
That December trade was not an isolated episode. Through the entire first half of 2026, the manager kept buying.
Corficolombiana — a Colombian financial group whose fixed income research is the reference on this market — identified 111 investment vehicles tied to the firm; 95 of them held TES as of June. Combined: 41.4 trillion pesos, roughly $13.6 billion. That is 26.8% of all Colombian local debt held by foreigners.
A year earlier, that share was 2.4%.
Between June 2025 and June 2026, total foreign holdings of TES grew by about $14.7 billion equivalent. Of that, some $12.7 billion was Pimco. Eighty-six percent.
But the number that stopped me was a different one
Those figures are large and eye-catching. This is the one that matters:
Between December and June, total foreign holdings fell by roughly $3.1 billion. Over the same period, Pimco’s position rose by roughly $5.6 billion.
Read that again, slowly.
Without that single manager, the foreign balance in Colombian debt would have dropped by close to $8.8 billion in six months. There was no wave of foreign capital arriving in Colombia. There was one enormous buyer covering everyone else’s exit.
Norway’s sovereign wealth fund cut its share from 8.8% to 2.5%. Franklin Templeton went from 3.5% to 1.7%. Dutch funds, Bluebay, Abu Dhabi’s sovereign fund: all selling. Corficolombiana described it as a stark divergence between one dominant buyer and a set of agents heading for the door.
An auction with one bidder
Picture the auction of a farm.
There are twenty people in the room. Bidding opens and the price climbs: 500, 600, 700 thousand. Those numbers being called out reflect what twenty people believe that farm is worth. That is a market price.
Now change the scene. Nineteen of them went quiet a while ago and are drifting out of the room. One man keeps raising his hand, and the auctioneer keeps calling higher numbers.
The price still rises. From outside it looks identical. But it is no longer the opinion of twenty people — it is the opinion of one. And if that man stands up and walks out, nobody is left holding that number up.
That is concentration risk. Colombia’s Banco de Bogotá put it more technically: the concentration reduced diversification and left the country’s assets vulnerable to the decisions of a single investor.
And in July, the man lowered his hand
This is where analysis turns into news.
In June, Pimco took light profits: net sales of about $330 million after the bonds rallied on the election outcome. Against a $13.6 billion position, that is barely 2%. It is not an exit.
But in July, foreign funds as a group became the largest net sellers in the market, with roughly $570 million of net disposals. It ended a three-month buying streak. Banco de Bogotá attributed it to profit taking.
This is not an exodus yet. It is a change of sign. And in a market held up by a handful of hands, the change of sign is the data point.
As I write this, the dollar has risen in Colombia for two sessions running: today’s official rate printed at 3,081.67 pesos, about twenty-five above yesterday. Don’t read too much into that. Against last week the dollar is still lower, and two days are not a trend. But after eight months of one-way traffic, it’s worth writing down.
So who bought what they sold?
Here is what I did not expect to find.
July’s largest net buyer of Colombian government bonds was the Finance Ministry itself, with about $2.1 billion — its heaviest month since July 2025, concentrated in 2027 maturities to reduce refinancing risk. Behind it, Colombian pension funds with a similar amount.
Sound familiar?
Two weeks ago we wrote about the U.S. Treasury doubling the buyback of its own bonds because the long end couldn’t find buyers. Different size, different country, same choreography: when the foreign buyer steps back, the issuer and domestic funds step in to hold the paper up.
This is neither conspiracy nor scandal. It is how sovereign debt markets behave when flows turn. But it’s worth recognizing when you’re watching it happen.
What to watch
If you would rather anticipate the turn than read about it three weeks later, here is what I would be tracking:
Monthly foreign holdings of TES. Published every month. Three consecutive months of net selling is no longer profit taking — that is an exit.
The rate differential. This carry trade lives on the gap between Colombia’s 12% policy rate and the Fed’s 3.50%–3.75%. Anything that narrows it — Colombian cuts, Fed hikes — drains the fuel. ING already argues the situation is screaming for a cut.
Moody’s. S&P has Colombia at BB− and Fitch at BB, both speculative grade. Moody’s is the only one still at investment grade, at Baa3, the final rung. If it falls below, many international funds are contractually obliged to sell regardless of what their manager thinks. That is the trigger that turns an orderly exit into a disorderly one.
The monthly chart. USDCOP is coming off a historic correction and is nowhere near equilibrium. Moves of this magnitude rarely end quietly.
What I’ll leave you with
The carry trade we walked through last week was real and it was profitable. Nobody cheated: rates were high, the currency appreciated, and some people saw it before the rest.
But every carry trade ends, and it ends quickly. It unwinds when the differential closes, or when somebody large decides they’ve made enough. In this particular case, “somebody large” has a name and has already started taking profits.
I don’t know when it turns. Nobody does, and be suspicious of anyone who tells you otherwise. What I do know is that a market with a single marginal buyer is a market that isn’t telling you the truth about what things are worth.
And that lesson travels far beyond the Colombian peso.
What are you seeing? Write to me below.
Join the discussion
No comments yet. Start the conversation.
