Argentina Mining Push-to-Talk Market: RIGI-Driven Demand, Size & Forecast 2026–2033

Market definition and scope

This report sizes what mining operators in Argentina spend on push-to-talk: the licences and platforms that carry the talkgroup, the handsets and consoles the workforce holds, the private radio and backhaul that make coverage possible where no public network reaches, and the integration and managed service wrapped around all three. Volume is users — individuals holding a push-to-talk identity on a mine's system. Value is end-user spend in current US dollars, not vendor revenue, and not the value of the minerals the mines produce.

The market exists because of a coincidence of geography and policy. Argentina's mineral provinces — Jujuy, Salta, Catamarca, San Juan, Santa Cruz — sit where public cellular coverage thins out or stops, so a mine cannot buy push-to-talk as a software subscription riding on somebody else's network. And from 2024 the RIGI large-investment regime began converting a pipeline of announced projects into approved investment plans with resolution numbers, which is the difference between a mining story and a procurement calendar.

Why a number this small decides something large

USD 11.18 million is a rounding error next to the capital it serves. Argentina's Ministry of Economy publishes the RIGI pipeline on a public dashboard — investment plans by sector and year, split by province, with mining broken into copper, lithium and gold and silver. Read for this report, the mining series runs 527 million in 2024, 1,473 in 2025, 2,579 in 2026, 3,875 in 2027, 4,128 in 2028, 3,278 in 2029, 2,450 in 2030, 1,503 in 2031, 416 in 2032 and 105 in 2033.

For the forecast window 2026–2033 that is USD 18,334 million of planned mining investment. Total push-to-talk spend across the same eight years is USD 198.71 million — 1.08 per cent of it. In the base year alone the ratio is 0.061 per cent, or 1,640 to one.

That ratio is this report's first argument. A procurement team will not spend a week on a line that is six hundredths of a per cent of a project's capital cost, which is precisely why the specification is usually inherited rather than chosen — carried over from the last site, or from whichever integrator answered the first email. The technical decision that determines whether a mine can talk to itself for the next decade is, in practice, often a decision made by nobody in particular.

What RIGI actually changes

The regime matters here less as a tax instrument than as a scheduling one. Its dashboard is explicit that a project is counted as approved only once it holds a resolution published in the Boletín Oficial. That makes the pipeline dated and public: a supplier can read next year's demand off a government website, which is unusual in any infrastructure market and close to unheard-of in mining communications.

The plan book also carries a composition shift that decides what kind of push-to-talk gets bought. Lithium outruns copper 1.8 to 1 across the 2024–2026 plan years. Copper outruns lithium 4.37 to 1 across 2027–2031. A brine operation and a copper mine do not buy the same system, and the difference is not the handset.

Five findings

1. Value grows faster than users, and price is not why. Value compounds at 19.95 per cent against 17.23 per cent for volume, but all-in realisation rises only 2.32 per cent a year. The gap is a mix effect: service revenue per user falls 10.6 per cent while radio and backhaul spend per user rises 44.1 per cent.
2. Coverage is the product, and its unit cost depends on which unit you use. Coverage capital per user is USD 2,600 at a brine field against USD 942.86 underground — a ratio of 2.76. Per square kilometre the same two sites run USD 7,222 against USD 110,000 — a ratio of 15.2. The two units point in opposite directions, and the choice between them is a business-case decision, not a technical one.
3. The largest block of new users is not a mine's payroll. Contractor and service fleets are not separately counted in 2026 and account for 20.7 per cent of 2033 users. They buy service rather than infrastructure, which is why the licence layer keeps a foothold even as owners self-build.
4. The buyer negotiates the cheapest line. Devices are 15.6 per cent of a site's capital; radio and backhaul are 63.4 per cent. The visible line item is not the consequential one.
5. Eighty per cent of the bill can be delivered domestically. Devices are 20.0 per cent of value and almost entirely imported; the remaining 80.0 per cent is survey, installation, integration and service labour that a local partner can supply. The regime's local-supplier register is therefore an addressable obligation, not an obstacle.

What would have to be true

Three things. At least four copper mines must reach operation inside the window, because the density that improves coverage economics arrives with them. The private-network architecture must keep winning at remote sites against a rented service, because a mine that rents does not buy radio. And per-user realisation must keep rising as video and data join voice on the same platform. If the first fails, 2033 is USD 30.60 million. If all three hold, USD 46.90 million.

Basis of estimate

Three classes of input sit behind every figure in this report, and they do not carry the same confidence.

• Published and cited. The RIGI dashboard's structure, its sectors, its mineral splits, its rule that approval requires a Boletín Oficial resolution, its local-supplier register, and the investment-plan series quoted above are read from the Ministry of Economy's own page. The existence and scope of Argentina's mining investment regime, and the Secretaría de Minería's publication of mining indicators, are read from the mining secretariat's own site.
• Stated but not re-verified this cycle. The regime's operating parameters — a reduced corporate income tax rate, a multi-decade stability guarantee, accelerated depreciation of capital goods, and relief from duties and VAT on imported capital equipment during the investment phase — are stated here as the operative terms of the establishing law. They were not re-read from the Boletín Oficial or the consolidated text for this report. Any figure that goes into a purchasing or financing decision should be confirmed against the consolidated instrument.
• Modelled. Subscriber counts, users per site, site counts by archetype, per-site capital stacks, all-in realisation per user and every segment share are Pheonix estimates. No Argentine statistical series publishes mining communications headcount or telecom spend at this granularity. The site archetypes are built from the operating and announced project list in the mineral provinces; the capital stacks are built from the four product layers described above.

Two exclusions matter. General IT and corporate connectivity are not counted, only the layers that carry push-to-talk. And no attempt is made to convert figures to constant dollars: all values are current nominal US dollars, so a reader comparing 2026 with 2033 should treat the growth rate as nominal.

Table des matières

• Competitive landscape: four layers, four different contests
• Value chain: what a site actually costs
• Investment activity: who funds a twenty-year asset
• Technology and innovation: where the coverage comes from
• Market risk: five risks ranked by durability
• Regulatory landscape: four instruments, one calendar

Paysage concurrentiel

Four layers, four different contests

Argentina’s mining push-to-talk market has no single competitive arena. It has four, contested by different companies with different economics. The table allocates the 2026 pool by layer and reports modelled concentration at each.

Read across the last two columns and the market’s central oddity appears. The most concentrated layer — radio and backhaul, where two suppliers take 64 per cent of a 4.28 million dollar pool — is 38.3 per cent of what the customer spends. The least concentrated layer is the one the customer negotiates with: integration, where the top three hold 21 per cent. The counterparty with the least market power sits between the buyer and the suppliers with the most.

That structure has a predictable consequence. The integrator earns its margin on labour and programme management, not on the radio, and its fastest route to a competitive bid is to specify the OEM that discounts hardest this quarter. The buyer, meanwhile, believes he has run a competitive process because he received three integration bids. He has run a competitive process on 10.2 per cent of his spend.

No supplier clears fifteen per cent

Because each layer has its own leaders and few companies lead in more than two, the combined pool is unusually unconcentrated. On our modelling no supplier, measured across all four layers, clears fifteen per cent of 2026 value — which is why this report records a tier-one supplier count of zero. The market is not consolidated; it is stacked. For the global context in which these four layers sit, see Pheonix’s global critical communication market coverage.

Hytera illustrates the stack from the hardware end. Its published portfolio spans TETRA and DMR radio, PoC handsets, MCS devices, a 4G/5G core with integrated base stations, satellite terminals, and — explicitly — a mining vertical, backed by a partner network it puts at more than 5,000 and an installed industry base it counts at 19 million users. That is three of the four layers from one supplier, and it is the closest thing this market has to a full-stack offer.

Nokia illustrates the opposite strategy: mission-critical enterprise networking with mining among its named verticals, and strength in the transport and core layers that carry the traffic rather than in the handset that keys the talkgroup. The two companies compete for the same site budget while barely competing for the same line item.

The moat is the file, not the box

What determines who wins a site is not the device specification but the qualification package: radio survey, spectrum authorisation, intrinsically-safe equipment schedule, safety case and integration with fleet management and identity. Nine to eighteen months of work, site-specific, largely undocumented outside the winning bidder’s own project folder.

This produces share that moves in steps rather than increments. A site that chooses an architecture is committed to it for the life of the radio — seven to ten years — and switching means repeating the entire file. Argentina’s installed base is therefore a series of islands, each dated by the year it was commissioned, and the supplier who wins a wave of commissioning wins a decade.

It also explains why the incumbent’s advantage is stronger than its technology. In a market where the largest supplier holds under fifteen per cent of the pool, the top two in radio hold 64 per cent of their layer — because in radio, the moat and the qualification file are the same object.

Where the channel earns its margin

Integration is 10.2 per cent of value and controls the specification of the other 89.8 per cent. That asymmetry is the most exploitable fact in this market for anyone who is not an OEM: a local integrator with eight mine relationships and no product to push is worth more per dollar of revenue than an OEM’s regional office, because the integrator decides what gets specified.

The corollary is a warning for buyers. In a market where no supplier is large enough to be accountable for the whole outcome, the only party who can be held to the integration risk is the one with the smallest balance sheet. Contract structure, not vendor choice, decides whether that risk is actually transferred.

Chaîne de valeur

What a site actually costs

A push-to-talk deployment is priced per site long before it is priced per user. The three archetypes below carry the whole model: a copper-scale operation with 1,400 users, a lithium or gold operation with 500, and an exploration or development camp with 60. Each figure is a modelled capital stack, not a quotation.

The spread across the three archetypes is only 1.30 times between cheapest and dearest per user, which is smaller than most operators expect. The spread in coverage cost per square kilometre is 15.2 times. The reconciliation of those two statements is density, and density is the one variable in this market that a mine cannot choose.

The two units point in opposite directions

Compare the two site types Argentina is actually building. A brine operation covers something like 180 square kilometres of salt flat with around 500 people on shift. An underground copper mine covers perhaps 12 kilometres of drift and decline with 1,400.

Underground costs fifteen times more per kilometre covered and roughly a third as much per user. Neither number is wrong. A network designed for a salt flat is a small number of high sites aimed across a large empty area; a network designed for a decline is a long linear emitter — leaky feeder or a chain of distributed antennas — serving a dense population along one axis. The salt flat’s problem is area. The mine’s problem is the cost of the first kilometre, not the thousandth user.

This is why the sequence of the RIGI plan book matters so much. Lithium, which dominates the 2024–2026 plan years 1.8 to 1 over copper, built the least favourable coverage economics in the country first. Copper, which dominates 2027–2031 by 4.37 to 1, brings four to five times the user density per covered kilometre. A supplier who read the 2025 market as a proxy for the 2030 market would misprice both.

Where the barrier to entry sits

Devices are 15.6 per cent of a lithium-scale site’s capital. Radio and backhaul are 63.4 per cent of it. The handset is the visible line and the least consequential: it is bought from a catalogue, it is replaced every three to five years, and its price is set by a global hardware market that Argentina does not influence.

What is not bought from a catalogue is the file. A site deployment requires a radio survey, a spectrum authorisation for the band it will use, an intrinsically-safe equipment schedule for the classified areas, a safety case tying radio coverage to emergency response times, and — increasingly — integration with fleet management, autonomous haulage and the mine’s identity provider. That package takes nine to eighteen months to assemble and is specific to one site. It is the reason a competitor with a cheaper box does not win next year’s mine.

The twenty-year arithmetic

Take the lithium-scale site. At a five-year amortisation its radio and backhaul layer costs 260,000 dollars a year; with service of 186,000 and support of 24,000, the all-in is 470,000, or USD 940 per user. Amortise the same capital over ten years and it falls to 130,000, taking the all-in to 340,000 — USD 680 per user — a 27.7 per cent reduction from a change in accounting horizon and nothing else.

That sensitivity is not an accounting curiosity. Radio and backhaul is a twenty-year asset being evaluated by a procurement team on a five-year business case, and the gap between those two horizons is where the architecture argument is actually won or lost. The regime’s stability guarantee exists precisely to make the long view rational for the owner; the supplier who prices on the short view is arguing against the instrument his customer has just signed.

Activité d'investissement

The path, not the total

The shape of the forecast matters more than its endpoint. Growth decelerates from 26.1 per cent in 2027 to 12.1 per cent in 2033 while the market nearly quadruples — the signature of a capital-led build rather than a demand-led one, tracking the RIGI plan book with a one- to two-year lag.

Peak growth in 2028–2029 sits one to two years after the mining plan series peaks in 2028, which is the lag you would expect between approval, construction and the commissioning of a communications system. A reader who wants an early signal about the 2031 push-to-talk market should watch the 2028 resolution numbers, not the mineral price.

What it takes to serve the increment

The base holds 9,660 users and the forecast ends at 29,400 — 19,740 net new. At a blended capital requirement of USD 3,400 per net new user, which reflects a mix that shifts toward denser copper sites, that is USD 67.12 million of site capital to serve the increment.

The radio and backhaul layer, at roughly 42 per cent of spend across the period, absorbs USD 78.76 million of the 187.53 million spent between 2027 and 2033. The difference — USD 11.64 million — is replacement and refresh of the installed base rather than growth. That reconciliation is worth stating explicitly, because it is the check that the subscriber model and the capital model are describing the same market.

Against the plan book, the whole radio layer is 0.50 per cent of the mining investment planned for 2027–2033. The market is not capital-constrained and never will be. The constraint is the qualification cycle described in the technology chapter, and the small number of people in the country who can write that file.

Who funds a twenty-year asset

Three structures compete. The owner builds and owns, capitalising the network and depreciating it over the life of the mine. A carrier builds and operates under a multi-year managed-service agreement, holding the asset on its own balance sheet. An OEM finances the infrastructure against a service commitment, which is a way of buying the specification as well as the equipment.

The choice is usually made on cash-flow grounds and then rationalised on technical ones. That is a mistake in this market for a specific reason: the asset’s service life is twenty years, the mine’s investment horizon under the regime is multi-decade, and the procurement team’s horizon is a capital cycle. Of the three structures, the one that aligns those horizons is the managed service — which is why service and platform stays above 24 per cent of realisation even as the market moves toward private infrastructure.

The procurement window

There is a timing asymmetry that a supplier should exploit and a buyer should notice. The regime’s capital reliefs — accelerated depreciation and duty relief on imported capital equipment — apply to spending inside the investment phase. A mine that defers its network to the operating phase buys the same equipment without the relief, in a currency it no longer has a stability guarantee over.

The practical consequence: networks specified after commissioning don’t get built. They get rented. The architecture decision, in other words, is often made by an accountant’s calendar rather than a radio engineer’s survey — and the forecast’s shift from 22 to 58 per cent private-network users is a bet that the calendar usually wins.

Technologie et innovation

Where the coverage comes from

Three technology families are in service in Argentina’s mining provinces, and the forecast is a migration between them rather than a replacement of one by another.

Two standards families sit behind that table. TETRA, P25 and DMR are narrowband land mobile radio systems with a two-decade record of availability; MCPTT and the wider mission-critical services family are 3GPP specifications that put push-to-talk on a broadband bearer alongside video and data. TCCA, the critical communications association, certifies mission-critical broadband services and runs the industry’s broadband transition programme — institutional evidence that the two families are converging into a hybrid rather than one displacing the other.

The 2026 composition is unusual by world standards, and the reason is the age of Argentina’s installed base. Gold and silver operations in Santa Cruz commissioned narrowband systems when narrowband was the only option, and those systems are still inside their service life. The forecast does not assume they are ripped out early. It assumes they are retired at end of life and not replaced in kind.

The physics of a salt flat

A brine field presents a radio engineer with three problems that have nothing to do with push-to-talk. There is nothing to mount equipment on, there is no grid to power it, and there is no fibre to backhaul it. Coverage therefore means a small number of high sites — masts, or whatever elevation the terrain offers — aimed across tens of kilometres, each fed by microwave or satellite.

The backhaul is where the technical argument usually goes wrong. A push-to-talk call setup is a signalling exchange: press, authenticate, join the talkgroup, speak. Push the core far enough away across a high-latency satellite hop and the first syllable arrives after the speaker has said it. Hosting the mission-critical core at the site — an edge deployment on the mine’s own infrastructure — takes the backhaul out of the signalling path and leaves it carrying only the media, where latency degrades quality rather than function.

That single architectural choice does more for the user experience than any handset specification, and it is invisible in a device tender. It is also, incidentally, why the radio layer’s share of realisation rises from 38.3 to 47.0 per cent over the forecast: the intelligence is moving onto the site.

Why autonomy makes the radio production-critical

Autonomous haulage, remote-controlled loaders and tele-remote drilling convert communications from a convenience into a dependency. Once a shovel is operated from a console two kilometres away, or a haul truck’s route is dispatched over the network, an outage stops tonnes rather than merely inconveniencing a shift.

This changes who owns the budget. In a conventional operation the radio system is bought by communications or IT and sits in operating expense. In an autonomous operation it is bought by operations, specified to an availability target, and treated as plant. The technology did not improve to make that happen — the operating model changed, and dragged the procurement with it.

What is not solved

Three things, stated plainly. Intrinsically safe broadband is the binding constraint underground: ATEX-rated handsets exist but are fewer, heavier and dearer than their non-certified equivalents, so the network can be installed before the fleet that would use it. Leaky feeder remains a narrowband solution carrying a broadband promise, which makes it a bridge rather than a destination. And power at remote sites — solar and battery at altitude, in cold, dust and thin air — constrains how much radio intelligence can be pushed to the edge.

None of these is a reason the forecast fails. Each of them is a reason the migration in the table above takes seven years rather than three, and a reason the legacy layer retains a fifth of the users in 2033.

Risque de marché

Five risks ranked by durability

Most risk registers rank by likelihood. This one ranks by durability, because in a market whose demand is a derivative of someone else’s capital programme, the risks that matter are the ones that persist rather than the ones that spike.

The first risk is the one that carries the forecast, and it is not a market risk at all. It is a project-finance risk with a communications consequence. The copper mines assumed in the 2033 base have not all reached a final investment decision, and a single deferral moves the national number by four to seven per cent — a large sensitivity for a factor that no supplier influences.

The loss that is not a loss

Lithium’s share of push-to-talk value falls from 42.9 per cent in 2026 to 28.1 per cent in 2033. That reads like a decline and is not one: lithium’s absolute spend rises from USD 4.80 million to USD 11.23 million, a multiple of 2.34. The share falls because copper grows faster, not because lithium shrinks.

This distinction matters commercially. A supplier who reads share loss as demand loss will under-resource the lithium installed base in exactly the years it is generating its most reliable revenue — brownfield expansion, additional shifts, contractor mobilisation — while chasing copper projects that will not commission until the end of the decade.

Scenarios

The band is asymmetric: the downside is 23.4 per cent below base, the upside 17.4 per cent above. That asymmetry is not a judgment about mining optimism; it is arithmetic. Slippage is a subtraction — a site that is not built contributes nothing — while upside requires several things to go right at once on the same site.

Concentration

There are roughly 45 push-to-talk sites in Argentina’s mining sector in 2026, and on our modelling the ten largest buying entities account for 61 per cent of value. That is not unusual for an industrial communications market, but it has a specific consequence for anyone forecasting from a pipeline: the national number is a sum of a small number of discrete decisions, so it behaves like a count rather than a trend.

Which is the honest limitation of this forecast. It does not describe a market that will grow smoothly at 19.95 per cent a year. It describes one that will grow in steps, each step being a commissioning, and the CAGR is simply what those steps average out to over seven years.

Paysage réglementaire

Four instruments, one calendar

Four regulatory layers touch a push-to-talk deployment in Argentina, and only one of them has a date a supplier can plan against.

The Ministry of Economy’s RIGI dashboard is explicit that a project counts as approved only when it holds a resolution published in the Boletín Oficial, and it publishes investment plans by sector, by year and by province alongside a register of local suppliers. For a communications supplier this is an unusual gift: the demand pipeline is published, dated and attributable to a named project in a named province.

Why the regime tilts the architecture

The regime’s incentives are weighted toward capital rather than toward operation. A reduced corporate income tax rate, a multi-decade stability guarantee covering tax, customs and regulatory terms, accelerated depreciation of capital goods and relief from duties and VAT on imported capital equipment during the investment phase all lower the cost of owning an asset. None of them lowers the cost of renting a service.

A mine inside the regime therefore faces a different relative price for a private network than a mine outside it — and that, more than any technical preference, is why the private-network share of users rises from 22 to 58 per cent in this forecast. The regime did not choose an architecture. It changed the price of two alternatives and let the site choose.

Verification note: these provisions are stated as the operative terms of the regime’s establishing law and were not re-read from the Boletín Oficial or the consolidated text for this report. See the basis of estimate in the market overview.

The local-supplier register and the eighty per cent

The dashboard maintains a register of local suppliers — companies the regime’s projects can draw on for goods and services. On our layer split, 20.0 per cent of the pool is devices and the remaining 80.0 per cent is survey, radio installation, integration, training and managed service, most of which can be delivered by an Argentine firm with the right certifications.

That ratio is the single most useful commercial fact in this report for a domestic integrator: the obligation that looks like a barrier is, in four-fifths of the spend, satisfiable from the buyer’s own province. The part that cannot be localised is the handset, and the handset is also the part with the lowest margin.

The instrument that is not a licence

One structural difference separates this market from a public-safety communications market, and it is worth being precise about. A mine’s push-to-talk system is not a licensed public service; it is private infrastructure on private land, subject to spectrum authorisation and to the safety obligations of the mine itself. Spectrum in Argentina is administered by ENACOM, the national communications regulator. That means the regulator’s leverage is at the point of authorisation, not at the point of operation — and the safety case, not the regulator, is what sets the coverage target.

Which is why coverage requirements at Argentine mine sites are typically written by the operator’s own risk function, in the language of emergency response times, rather than inherited from a national standard. A supplier who can read a safety case is at a structural advantage over one who can only read a datasheet.