On 21 August 2026 the central bank was publishing, in one table and for one date, four official exchange rate references, two of them the exchange rate bands: 744,41 · 1 497,4528 · 1 518,13 and 1 868,54 pesos per dollar. A commercial lease agreed in dollars is collected in pesos, and the clause choosing among those references decides the amount before any negotiation over the rent. What follows is not a forecast of the exchange rate but three scenarios defined by what each of them needs in order to happen.
Exchange rate bands and two communications: four official references
The principal indicators of the BCRA publish each row with its own date. As at 21 August 2026 the exchange rate band regime had an upper limit of 1 868,54 pesos per dollar and a lower limit of 744,41; the retail rate, published under Communication B 9791 as a seller average, was 1 518,13; and the wholesale reference, under Communication A 3500, 1 497,4528.
| Reference | Value at 21 August 2026 | Published under |
|---|---|---|
| Exchange rate band, lower limit | 744,41 | Exchange rate band regime |
| Wholesale reference rate | 1 497,4528 | Communication A 3500 |
| Retail rate, seller average | 1 518,13 | Communication B 9791 |
| Exchange rate band, upper limit | 1 868,54 | Exchange rate band regime |
The distance between the retail and the wholesale rate is 1,38 %. The distance between the wholesale rate and the upper limit is 19,86 %, and between the wholesale rate and the lower limit, a factor of 2,01. The whole band measures 1 124,13 pesos per dollar, or a factor of 2,51 between its two ends. A contract saying “in dollars” without naming the reference leaves open a difference of that size.
The gap between retail and wholesale deserves a reading of its own because it is not a market datum but a circuit one. The retail rate published under Communication B 9791 is a seller average of institutions, and the wholesale rate of Communication A 3500 is a reference constructed by the central bank itself. A contract settled by bank transfer in pesos and one settled at an institution’s counter do not use the same figure even though both say “official dollar”. At 21 August 2026 that difference was 1,38 %, which over an annual rent is smaller than the usual error of a negotiation, and which over a five-year contract stops being so if the gap widens.
The fork: what decides the peso value of the rent
The starting point is the same for all three scenarios. The rent is expressed in dollars, payment is made in pesos, the conversion uses a published reference, and the tenant invoices in pesos. What changes between scenarios is not the rent in dollars but the relation between that conversion and internal prices, that is between the exchange rate and inflation.
The magnitudes defining that relation are in the same table. Monthly inflation at 31 July 2026 was 2,1 %, the annual figure 33,8 %, and the median of expectations of the Market Expectations Survey for the following twelve months, 21,8 %. Repeating July’s monthly pace twelve times would give 28,3 % a year, which is an arithmetic operation on a published figure and not a projection. The central bank’s international reserves were 50 006 million dollars at 19 August 2026, a figure the table itself calls provisional and subject to changes in valuation.
With those elements, the fork comes down to one question: whether the conversion rate rises faster, at the same pace, or more slowly than internal prices during the life of the contract. Each answer defines a scenario, and each scenario has an observable trigger.
Scenario A: the wide corridor
Condition of occurrence. The wholesale rate stays inside the band without approaching either end, and annual inflation descends from 33,8 % towards the median expectation of 21,8 %.
What it implies for the contract. The rent in dollars converted into pesos grows more slowly than internal prices. For a tenant invoicing in pesos, the relative weight of the rent within its costs falls without anyone renegotiating anything. For the owner, income in dollars holds and income in pesos loses purchasing power against its own costs if those follow inflation.
Where it shows first. In the distance between the wholesale rate and the upper limit, today 19,86 %. While that distance does not narrow in a sustained way, the scenario holds. The second signal is the gap between published annual inflation and the survey median: today 12,0 percentage points.
Scenario B: the ceiling as the operative reference
Condition of occurrence. The wholesale rate approaches the published upper limit and the distance of 19,86 % narrows in a sustained way, with annual inflation not descending at the pace the median expectation suggests.
What it implies for the contract. The conversion of the rent grows faster than the tenant’s internal prices. Rent gains weight within the cost structure of the occupier, and renegotiation pressure appears before expiry, even with the dollar rent untouched. For the owner, income in pesos improves in relative terms, but the occupier’s default risk grows for the same reason.
Where it shows first. In the row of the upper limit and in that of the wholesale rate, both published daily and with the same date. The relation between the two is the most direct indicator: today the wholesale rate stands at 80,14 % of the ceiling. The second signal is the gap between retail and wholesale, today 1,38 %: a sustained widening of that gap changes the conversion cost for whoever pays through the retail circuit.
What it does not imply. This scenario says nothing about the level of the rent in dollars nor about demand for space. A contract signed in a market with high vacancy and one signed with low vacancy behave identically in the face of the same conversion movement, because the clause operates on an amount already agreed. Confusing the two layers — what is negotiated and what is converted — is what produces renegotiations that resolve nothing, because they attack the rent when the problem is in the date and the conversion reference.
Scenario C: the band is redefined
Condition of occurrence. The published values of the limits change level, or the regime stops being published with those two rows. This scenario does not depend on the market but on an administrative decision, and its trigger is documentary rather than statistical.
What it implies for the contract. Any clause referring to a named reference — the upper limit, the lower limit, or either of the two communications — ends up pointing at a different definition or at a row that no longer exists. The contract does not become invalid, but its conversion mechanism comes to depend on an interpretation, which is exactly what the clause was meant to avoid.
Where it shows first. In the indicator table itself, which shows the names of the rows before their values. The disappearance or renaming of a row is visible the same day. This text did not verify the norm establishing the limits, so it does not state whether they are movable, fixed or revisable, nor with what periodicity.
Which indicator shows which scenario
| Indicator | Reference value | Which scenario it signals if it moves |
|---|---|---|
| Wholesale over upper limit | 80,14 % at 21 August 2026 | Rising in a sustained way, scenario B |
| Distance from wholesale to upper limit | 19,86 % | Narrowing, scenario B; holding, scenario A |
| Retail against wholesale gap | 1,38 % | Widening, it raises the conversion cost in the retail circuit |
| Annual inflation against REM median | 33,8 % against 21,8 % | Converging, scenario A; the gap holding, scenario B |
| Names of the band rows | Two rows published daily | Changing or disappearing, scenario C |
| International reserves | 50 006 million dollars at 19 August 2026 | — |
The last row of the table carries a dash on purpose. Reserves appear in the same table and at the same frequency, but no reserve threshold is associated in that publication with a change of band, and setting one independently would be inventing a rule. They are included because they form part of the same dashboard, not because they trigger a scenario.
What it means for the contract
The practical consequence is not choosing a scenario but drafting the conversion clause so that all three are legible. That means three concrete and verifiable decisions. The first is naming the reference: not “dollar” but, for example, the wholesale reference rate of Communication A 3500, with the value of a date defined in relation to the payment due date.
The second is fixing the conversion date with the same precision: the due date, the preceding business day, or an average of defined days. Between 21 August 2026 and any other day there is a difference neither party controls, and leaving it undefined transfers that difference to whoever has less bargaining power at the moment of payment.
The third is providing for what happens if the chosen reference stops being published. A substitution clause naming a second official reference — and only an official one — turns scenario C from a problem into a formality. Without it, the disappearance of a row in a central bank table becomes an argument about the will of the parties years after signature.
It is also worth separating what this text verified from what it did not. It verified the four values, their dates and the communications under which they are published, all in a single central bank table consulted on 23 August 2026. It did not verify the norm establishing the band limits, nor its updating regime, nor whether the published values move daily under a written rule or stay fixed until a decision. That is why scenario C is defined by what is observed in the publication and not by what an administrative act, unread here, would provide.
Conclusions
As at 21 August 2026 the central bank was publishing four official references for the dollar: a lower band limit of 744,41, a wholesale reference under Communication A 3500 of 1 497,4528, a retail seller average under Communication B 9791 of 1 518,13, and an upper band limit of 1 868,54 pesos per dollar. Between the first two there is a factor of 2,01; between the wholesale rate and the ceiling, 19,86 %; between retail and wholesale, 1,38 %. With monthly inflation of 2,1 % and annual inflation of 33,8 % at 31 July 2026 against a median expectation of 21,8 % over twelve months, the three scenarios described are distinguished by triggers observable in the same table and not by probabilities, which are not assigned here. Scenario A is recognised by a distance to the ceiling that does not narrow; B, by a distance that narrows with inflation that does not converge; C, by a change in the names of the rows rather than in their values. None of the three alters the rent agreed in dollars: they alter its peso equivalent, which is what is actually collected and paid. The operative conclusion is therefore one of drafting and not of market: the clause naming reference, date and substitute is worth more, in any of the three scenarios, than any correct call on where the exchange rate is going.
Este material tiene carácter informativo y no constituye asesoramiento legal, fiscal ni financiero. No se cita ninguna renta de mercado, ninguna vacancia y ninguna norma que establezca los límites de las bandas cambiarias: esas cifras y ese acto no fueron verificados en fuentes primarias para este texto, y las bandas se mencionan únicamente como indicadores publicados por el Banco Central.
