A 5G future ahead

If there were ever a book written on Karachi’s bazaars, Empress Market would undoubtedly feature on the cover. Perhaps the most interesting part of the shopping experience is the once-live auctions conducted outside: a crowd gathers around a man shouting into a loudspeaker, bids called out in the open — striking just the right balance between amusing and shady.

That scene may have faded since the anti-encroachment campaign, but the format has found its way to grander stages lately.

The Pakistan International Airlines sale, years in the making, finally concluded with the dramatic flair that only live bidding can produce. The Pakistan Super League switched from quiet drafts to open auctions, turning franchise decisions into appointment viewing.

But arguably the single-most meaningful auction of the past decade took place on March 10, even if there was no spectacle as such. On that day, the telecom industry finally sat down to bid for new spectrum, a demand it had demanded for years.

At roughly 750 MHz post-auction, Pakistan has crossed the lower-middle-income country average for the first time

What followed was 480 MHz being sold, more than doubling Pakistan’s existing spectrum of just 274 MHz, which itself was among the lowest in the region and income group. In the process, the government also managed to raise $507 million.

Radio spectrum — the invisible resource over which wireless signals travel — comes in different frequency ranges, each with different physical properties. Low-frequency bands (like 700 MHz) travel farther and penetrate walls more easily, which makes them valuable for rural coverage. Higher frequency bands (like 3,500 MHz) cannot travel as far but can carry far more data in a dense area, which is why they are the global standard for 5G deployments.

The Pakistan Telecom Authority had put six frequency bands up for auction. Two of them — 1,800 MHz and 2,100 MHz — are legacy bands with existing infrastructure, and attracted no bids.

The four new bands — 700 MHz, 2,300 MHz, 2,600 MHz, and 3,500 MHz — are where the action was. Pricing ranged from $32.5m per lot for 700 MHz down to $6.5m per lot for 3,500 MHz, which reflects the denser and costlier site networks required for higher-frequency deployment rather than a lack of strategic importance.

Most of the available spectrum on offer was on the higher frequency bands — 190 MHz for 2,600 MHz and 280 MHz for 3,500 MHz, both unpaired. Naturally, given the regulatory push towards 5G, this is where bidding was also concentrated, as we saw the allocation of 190 MHz in the former and 220 MHz in the latter.

Understandably, telco-wise allocations are more interesting and give a glimpse of what their strategies could be in the coming years. Jazz spent the most money — $239.5m — and bought the broadest portfolio, the only operator across all four bands.

The defining purchase was for 700 MHz, for which Jazz was the sole bidder. “Securing spectrum in the 700 MHz band is pivotal to closing Pakistan’s digital divide”, said CEO Aamir Ibrahim, “as it enables us to extend reliable, high-quality connectivity across vast geographies, particularly in rural and underserved communities”.

The wide mix is consistent with Jazz’s market dominance, where it commands an almost 37 per cent share in cellular and mobile broadband, each with 74m and 56m subscribers, respectively. In a separate press release, the company announced a $1 billion investment over three years to expand network capacity and modernise infrastructure in support of its 5G rollout.

5G’s speed promises are only as good as the backhaul connecting those towers to the core network; currently, only 15pc of Pakistan’s cell sites are connected via fibre

Ufone’s decision is also quite interesting; it emerged as the largest acquirer in the 3,500 MHz band with 120 MHz allocation. In retrospect, one can obviously rationalise it within the context of its merger with Telenor, which together boast some 49m broadband subscribers as of 2025.

Meanwhile, Zong bought the least and spent the least. Its portfolio — 60 MHz of 2,600 MHz and 50 MHz of 3,500 MHz — is quite lean. This is at least a little surprising since the telco boasts the highest mobile data consumption per subscriber in the local market by far, at 118 GB in FY25; no one else is in triple digits.

At roughly 750 MHz post-auction, Pakistan has crossed the lower-middle-income country average for the first time, though it only slightly trails the upper-middle-income group. But the country has, at minimum, stopped being an outlier.

Now that the question of spectrum availability has finally been resolved, after many pains and deliberations, the more difficult challenge is yet to come. The information memorandum sets ambitious targets: 1,000 sites annually, of which 200 are supposed to be greenfield builds.

However, towers alone are not enough: 5G’s speed promises are only as good as the backhaul connecting those towers to the core network. Currently, only 15pc of Pakistan’s cell sites are connected via fibre, and the International Monetary Fund requires operators to raise that to 35pc by 2032. For context, fiberising a single site in Pakistan costs anywhere between $10,000 to $20,000, and with tens of thousands of sites needing upgrades, the capital commitment extends well beyond what was spent at auction.

We also face a mounting demand-side problem where device accessibility remains a big question mark, as 5G phones typically cost upwards of Rs60,000-70,000. Luckily, the government’s mobile manufacturing policy is just around the corner and will hopefully spur things.

The most pressing concern probably remains the payback period. Pakistan’s average revenue per user for mobiles sits at approximately $1.10 per month, which is far below the $8.20 global average and among the lowest globally. However, superior networks could enable new monetisation strategies.

GSMA analysis shows that speed-based tariffs, where operators can charge premiums for higher speeds rather than data volume, correlate with faster monetisation growth across multiple markets.

It is, in other words, a solvable problem, just not one that spectrum alone can solve. To borrow from the oft-used analogy, while the road is now wider, the cars are still expensive. And the traffic, as anyone on Shahrah-e-Faisal will tell you, does not sort itself out on its own.

Mutaher Khan is co-founder of Data Darbar and works for the Karachi School of Business and Leadership

Published in Dawn, The Business and Finance Weekly, March 16th, 2026