Ever since the 1990s, telecom equipment vendors have had an unshakeable conviction in the network spending cycle. It should work something like this: A new and much-hyped technology appears, triggering a wave of investment by telcos scared of losing out to faster-moving rivals. Once these state-of-the-art networks have extended their reach, however, spending levels dip. The low point arrives when telcos are left making inexpensive tweaks to their systems, squeezing the last remaining bandwidth out of their pipes before the next generation comes along.

This is supposedly the phase we are in now, with high-speed 4G mobile networks widely available in developed markets but 5G still in the lab. That would explain why equipment makers like Ericsson AB (Nasdaq: ERIC) and Nokia Corp. (NYSE: NOK) are in such a rut: Listen to any of the quarterly earnings updates from those companies over the last year and one can hear senior executives repeatedly complain about the "weak spending environment" for mobile broadband equipment. A boilerplate statement in Nokia's financial reports emphasizes its dependence on "the cyclicality and variability of the information and telecommunications industries." (See Loss-Making Ericsson Still Short on Vision and Nokia Upbeat on Turnaround Despite Sales Decline.)

While there can be no doubt that service providers have slashed spending on mobile broadband rollout (if not other networks), 5G seems unlikely to deliver the same kind of cyclical bounce as its predecessors. That will have big ramifications for suppliers, and partly explains why Nokia thinks most of its growth in the next five years -- and possibly beyond -- will come from entirely different markets. (See Nokia to Create Standalone Software Biz, Target New Verticals.)

For one thing, previous generations of network technology may hold few pertinent lessons for 5G. The rollout of 3G in the early years of the millennium necessitated huge investments in new telecom sites and hardware to meet particular spectrum needs. When it comes to 5G, however, many of Europe's operators seem likely to use sub-GHz and mid-band frequencies that allow the technology to be "overlaid" on current sites -- much as a household architect might build modifications on an existing structure. (See DT, EE, Orange, Vodafone Open Up on 5G.)

Want to know more about 5G? Check out our dedicated 5G content channel
here on Light Reading.

Some kind of boost might come from investment in so-called "millimeter wave" technology, which uses very high frequency bands to deliver ultra-fast connections. Because signals travel over short distances in these ranges, many more sites will be needed to support services. Yet millimeter wave deployments will probably be concentrated in specific zones, such as busy airports, rather than blanket entire regions.

Networks have also become far more software-based in the last 15 years, allowing operators to make upgrades more easily and cost-effectively than when they were moving from 2G to 3G. Indeed, a number of technologies that vendors are selling today -- during the supposed low point of the network spending cycle -- are designed specifically to smooth the future transition to 5G. Ericsson, for example, is touting a radio platform it simply calls Ericsson Radio System (ERS) as a critical step on the 5G journey. Designed partly to facilitate software-based network upgrades in future, ERS accounted for about 15% of the total deliveries of radio units in 2016. Ericsson expects the figure to hit 50% this year.

But even if 5G does call for a sharp increase in capital expenditure, most service providers will probably see little need to deploy the technology rapidly and at scale. A few pioneers are moving quickly in the hope of influencing the eventual shape of the standard (risking some bifurcation, according to analysts) and for bragging rights. Others in markets with poor broadband competition see an opportunity to use 5G as a fixed-wireless access technology. But the cold reality is that no mobile applications currently exist that an advanced 4G network cannot handle. (See 5G on Track but Fragmentation Still a Concern and 4.5G Sets High Bar for 5G.)

Despite hype surrounding the Internet of Things, not to mention augmented and virtual reality, it is hard to see that changing in the next few years, especially as gigabit-speed and low-latency "4.5G" networks begin to emerge. "There is no service today that requires 5G as such and you will need a killer app before operators will invest massively in it," says Raphaël Glatt, the head of signaling for BICS , the international carrier division of Belgian telco incumbent Proximus .

Next page: An adjunct to 4G

An adjunct to 4G
All of this shines an interesting light on Nokia's 4.5G strategy. Perhaps more than any other supplier, the Finnish company has presented spending on more advanced 4G technologies as a necessary step in the "evolution" toward 5G. This does not in any way mean their rollout is a kind of "5G-lite" move. But it does imply that operators will not face the need for a dramatic 5G overhaul provided they continue to invest in 4G. (See Nokia Boasts 4.5G Momentum With 110 Deals.)

Nokia's expectations about the nature of 5G deployment seem to back this up. Phil Twist, the vice president of marketing and communications for the Finnish vendor's mobile business, says there are two ways of building 5G networks, and that both make it very different from 3G and 4G. In the first scenario, he explains, an operator would roll out 5G as an "extension" to an existing network, using a cloud packet core and high-bandwidth connectivity elements that Nokia is starting to promote today.

"There is nothing to stop you adding 5G as just another radio interface, but there is a second model that some people are considering, which is a standalone 5G network," he says. "You put in a 5G access point and dedicated 5G core network covering a specific use case in a specific geography and then evolve that over time."

The point is that in each case 5G does not replace 4G in the same way the latter quickly made 3G look obsolete. "For a considerable length of time, 5G is going to be an adjunct to a wide area LTE network," says Twist.

Moreover, Nokia's own forecasts suggest the company is expecting a fairly subdued 5G party following standardization, with limited binging on the technology in its early days. During its capital markets day last November, Nokia predicted that its main addressable market (that is, the telco opportunity) would increase at a compound annual growth rate of just 1% over the subsequent five years. Given that a standardized version of 5G is expected to hit the market in 2020, this forecast obviously covers the period when 5G is first likely to see commercial deployment. And while the 2.2% market decline Nokia anticipates this year could bode well for growth toward the end of the forecast period, the increase seems unlikely to bedazzle.

Twist says it would be "dangerous" to make a long-term prediction about where markets are going but sounds downbeat. "Of course we would like the market not to be flat but that is why we are looking at more than connectivity," he says, referring to Nokia's new strategy of expanding into a range of "adjacent" vertical markets and doubling down on its still-small software business.

Rival Ericsson, however, is even less forthcoming. At its own investor update in November, the Swedish supplier said the market for radio access network equipment -- which accounts for most of its sales -- would probably shrink by 2-6% this year, after declining by 10-15% in 2016. It refuses to provide any kind of longer-term outlook but remains beset by strategic uncertainty following a sequence of earnings disappointments, failed restructuring efforts and recent management changes.

Ericsson is expected to have plenty to say about 5G at next week's Mobile World Congress. But if it chooses to present 5G as the next big upswing in the time-honored network spending cycle, it will have plenty of convincing to do.

— Iain Morris, , News Editor, Light Reading