Limestone & Silicon - what Taiwan did that Malta didn't

Malta and Taiwan, two islands and the roads not taken. Two small islands set out from poverty at roughly the same hour. One built the machine the world now cannot live without. The other sold its passport.

For a prettier, more interactive version of this article, please click through here.

Before we begin — a few caveats

I have never set foot in Taiwan. I am not an economist, I am not a political scientist, and I am not a historian.

What I am is an interested amateur with a keen and abiding curiosity about all four of those things — Taiwan, economics, political science and history — and about the small island I happen to live on. This is an essay, not a peer-reviewed paper. All opinions are mine and are clearly marked as such. Read the data as data and feel free to dispute it, and read the argument as just this one guy’s argument, and feel free to disagree with me and counterargue.

In the interest of full disclosure: I used Claude to help in research, initial rough draft, and formatting and presentation of this piece.

The arguments and the opinions — and any errors — are mine alone. Don’t blame the bot.

A podcast, an island, and an uncomfortable mirror

This piece began with an episode of the Search Engine podcast called “The Many Lives of Taiwan.”

In it, host PJ Vogt and his guest — the political scientist Shelley Rigger, the Brown Professor of East Asian Politics at Davidson College and the author of Why Taiwan Matters: Small Island, Global Powerhouse and The Tiger Leading the Dragon — trace how a tiny, threatened island escaped demise. They follow Taiwan from colonial subjugation, through an era of mass Barbie, plastic-toy and cheap electronic trinket production, and into its current role as the technological powerhouse on which the entire modern world depends.

I listened to it as someone who lives on a different small island. And I could not shake the feeling that I was looking into an uncomfortable mirror — that Taiwan’s story is, in part, a story of what could have been for Malta. Two specks of rock in contested seas. Two former colonies that started their independent economic lives within a few years of one another, both poor, both resource-starved, both forced to make a living from human ingenuity because they had little else.

One of them chose, deliberately and patiently, to become extraordinarily good at a few extraordinarily hard things.

The other, I will argue, chased a series of easier ones.

What follows is first the evidence, laid out as plainly as I can manage — and then the argument.

The ledger — the two islands at a glance

Before any argument, the facts. Here is how the two islands compare across the measures that matter — size, people, output and the distribution of it.

A note on honesty: by the headline measure most people reach for — GDP per head — Malta is not a poor country. It is wealthy. The interesting divergence is not in the size of the number but in what produced it, and in how durable it is likely to prove.

Measure	Malta	Taiwan
Land area	316 km² — smaller than many cities	≈ 36,200 km² — about 115× Malta [3]
Population	≈ 574,000 — end of 2024 [7]	≈ 23.2 million — about 40× Malta [6]
Population trajectory	Growth — but borrowed. The headline +1.9% in 2024 was almost entirely immigration; the natural increase (births minus deaths) collapsed to just 193 people, and over 2014–24 Maltese-family births added barely ~5,660 while foreign residents grew by ~131,000 [8]	Flat to falling. Annual growth ≈ 0.2%; an ageing, low-fertility society [11]
Native birth rate	≈ 1.1–1.5 total fertility rate — far below the 2.1 replacement level, and below it since around 1990; the finance minister has called the decline “the greatest challenge of our time” [9]	≈ 0.9–1.1 — also far below replacement; low fertility is a shared affliction [11]
GDP (nominal)	≈ $23 bn — 2024 [13]	≈ $884 bn — 2025 [12]
GDP per capita (nominal)	≈ $41,700 — IMF, 2024 [14]	≈ $34,400 → $39,500 — IMF: ~$34.4k in early 2025, lifted toward ~$39.5k by the AI-chip boom [15]
GDP per capita (PPP)	≈ $60,500–67,700 — World Bank / IMF, 2024 [16]	≈ $76,000–98,000 — IMF estimates, 2024–26; comparable to Denmark or the Netherlands [17]
Industry share of the economy	≈ 10–11% manufacturing; the economy is ~87% services [18]	≈ 36% industry — a genuinely productive base [19]
Income inequality (Gini)	≈ 28–31 — Eurostat; relatively equal, but trending upward [20]	≈ 33.9 — 2023 est.; measured differently, so compare loosely [21]
World Happiness Report	≈ 40th — score ~6.35; Maltese under-30s ranked the unhappiest youth in the EU [22]	27th — the happiest place in East Asia, ahead of Singapore [23]
R&D spending (% of GDP)	≈ 0.5% — among the very lowest in the EU [24]	≈ 4.0% — third-highest on Earth, behind only Israel and South Korea [25]
Patent applications (annual)	≈ 140 to the European Patent Office (2017 peak) — and some of even those are filed by foreign IP-holding companies domiciled in Malta, not Maltese invention [34]	≈ 50,000 invention-patent applications filed at home each year — remarkably steady; even per head, on the order of nine times Malta’s rate [34]
The flagship	iGaming & finance — online gambling alone ≈ 12% of GDP; manufacturing led by foreign-owned plants [28]	TSMC — one firm makes >90% of the world’s most advanced chips [27]

Figures are the most recent available at the time of writing (2024–2026) and are drawn from the IMF, World Bank, Eurostat, the Asian Development Bank, the Taiwan and European patent offices, national statistics offices and the World Happiness Report. Where two bodies measure the same thing differently — as with the Gini coefficient, or with patents filed at different patent offices — both ranges are shown, and compared loosely, rather than with a false precision.

Four figures worth holding onto:

$154 — Taiwan’s GDP per head in 1951, among the poorest places on the planet. Today it is north of $34,000. [29]
90% — the share of the world’s most advanced semiconductors made by a single Taiwanese company, TSMC. [27]
$24.6 bn — the value of the slice of TSMC the Taiwanese state still holds, from a 1987 seed of perhaps $100 million. [33]
€1.4 bn — raised by Malta selling citizenship since 2015, a scheme an EU court has now ruled illegal. [26]

Island one — Taiwan: the improbable ascent

In 1949, the defeated Republic of China government retreated across the strait to Taiwan, bringing some two million people to an island that was a war-damaged agricultural society. Per-capita income sat at roughly $154 a year. Eighty-five per cent of the population worked the land. By any measure of the time, this was a poor, crowded, frightened place with a hostile giant ninety miles away. [29]

What happened next is the part worth dwelling on, because it was not luck. In the 1950s the government carried out genuine land reform — programmes with names like “Land to the Tiller” — which freed up both capital and labour for industry. [29] In the 1960s Taiwan opened the world’s first modern export-processing zone, in Kaohsiung, and put its cheap, increasingly literate labour to work assembling textiles, plastics and toys for export. This is the Barbie era the podcast describes: “Made in Taiwan” as a byword for cheapness, not technical sophistication. [30]

The decisive move came when that era was still working. Rather than ride low-cost assembly until the wage advantage evaporated — which it eventually always does, as is increasingly happening in mainland China now, where coastal manufacturing wages have climbed to several times those of Vietnam or Bangladesh and labour-intensive production is steadily moving on [45] — Taiwan’s technocrats chose to climb. In 1973, at the urging of figures like K. T. Li, often called the father of the Taiwanese miracle, the government founded the Industrial Technology Research Institute. Three years later it negotiated a now-legendary technology-transfer deal with the American firm RCA, acquiring chip-making process knowledge and training a generation of engineers in it. [31]

And then came the single most important act of industrial strategy of the late twentieth century. Taiwan did not try to beat Japan and the United States at the whole semiconductor business — product creation, design, branding, marketing and all. It looked honestly at what it could realistically dominate and picked the hardest, most operationally demanding link in the chain: precision manufacturing. In 1987, ITRI’s president Morris Chang spun out the Taiwan Semiconductor Manufacturing Company, founded with the Taiwanese state putting up roughly half of the start-up capital. [32]

Consider what that seed became. The government has long since sold down most of its holding, yet the National Development Fund — the state’s investment arm — still owns about 6.4% of TSMC and remains its single largest shareholder. That retained sliver alone was worth around US$24.6 billion at the end of 2022, and considerably more since, as TSMC grew into one of the most valuable companies on Earth. As the hosts of the Acquired podcast lay out in their history of the company, a state investment of perhaps US$100 million in 1987 became one of the most successful public investments in modern economic history. [33]

That same year, Taiwan lifted martial law and began its turn toward democracy. The two facts are not unrelated: a confident, rising society can afford to liberalise. Today TSMC produces more than 90% of the world’s most advanced chips, accounts for roughly 8% of Taiwan’s entire economy and 12% of its exports, and is so strategically vital that analysts call it the island’s “silicon shield” — the thing that makes the wider world need Taiwan to remain Taiwan. [27] Beneath that one famous name sit some 1.7 million small and medium enterprises, many of them global champions of one specific precision component. [30]

The lesson I take from Rigger’s account is not “build a chip industry.” It is something more transferable: pick a hard thing, commit across decades and across changes of government, and climb the value chain before the cheap floor falls out from under you.

Both islands were once poor. Both had to live by their wits. One spent fifty years getting deliberately, patiently good at something difficult.

Island two — Malta: the comfortable drift

Malta’s starting point was, if anything, more precarious. For a century and a half its economy was essentially a British naval base with an island attached; the Royal Naval Dockyard was the single largest employer. When Britain began winding down its military presence after the 1950s, and especially after independence in 1964, the island faced an existential question of how to earn a living at all. [35]

It is worth pausing on that dockyard, because it sets the pattern. The work was genuinely skilled — ship repair is precise, technical, demanding. But it was also mobile. Ship repair is a price-competitive trade, and over the following decades it sailed or drifted, as it always would, to cheaper yards: to Turkey, Bangladesh, the Far East and beyond. This is the recurring shape of the Maltese economy across its modern history — skilled work, certainly, but rarely work that could not pack up and leave.

For a while, the answer to the post-war question rhymed with Taiwan’s. From the 1960s, Malta deliberately positioned itself as a low-cost manufacturing base for foreign firms, building industrial estates and offering incentives. It worked. The German toy-maker Playmobil arrived in the mid-1970s and — remarkably — still makes essentially every Playmobil figure in the world at its Maltese factory, one of the island’s largest, having moulded billions of the little smiling people. [36] In the 1980s came a real push into textiles, plastics and consumer goods; in the early 1990s, engineering and electronics, anchored by the chip-packaging operations of ST Microelectronics, which remains one of Malta’s largest exporters to this day. [37]

Here is the counterfactual that haunts me. The textile and light-electronics push of the 1980s and early 1990s was a genuine fork in the road. Malta could have done what Taiwan did: treated low-cost assembly as a school rather than a destination. It could have used those decades to build deep, specific technical competence in two or three narrow fields — and then, as its own wages rose, kept the high-skill design, tooling and management on the island while offshoring the low-margin assembly to cheaper, stable neighbours: the Tunisian and Moroccan coasts, or mainland Sicily. (Libya, despite its proximity, was never a realistic partner — first Gaddafi, then the chronic instability that followed his fall.) The geography for a Maltese “headquarters economy” of specialised manufacturing was, and is, right there.

That is not, mostly, what happened. Global market shifts in the late 1990s and 2000s meant that most textile mass manufacturing moved to Asian countries with lower production costs.

Contemporaneously, Malta developed — and then, in my view, over-developed and over-exploited — a service economy. Tourism first: a real and natural endowment, now strained to the point that some three million visitors a year crowd an island of barely half a million. Then financial services, insurance, fund administration, company and intellectual-property domiciliation, and the dense thicket of corporate-service providers that grows up around all of it. For an island this small, building these clusters was not unreasonable; in some ways it was clever. But Malta was always going to be competing on this terrain — with Luxembourg, for example, and then with Ireland, and then with shadier, hungrier rivals like Cyprus.

It is a crowded race to a narrowing margin.

The headline case is iGaming. Malta hosts a remarkable share of Europe’s online gambling industry, worth something like 12% of GDP. [28] But notice what Malta actually owns in that arrangement: very little. The intellectual property, the platforms, the brands belong, overwhelmingly, to foreign firms. Malta supplies the licence, the office space, the regulatory wrapper and the labour. It is a service provider, not a creator — and a service provider can always be switched.

EU accession in 2004 and the euro in 2008 then opened the door to something faster and easier still than even services: regulatory arbitrage. Malta discovered it could sell access directly. Access, from 2014, to EU citizenship itself, through a passport-sale scheme. [39] And from 2018, a much-trumpeted bid to become “Blockchain Island,” courting the world’s cryptocurrency exchanges with a bespoke legal framework. [40]

Underpinning the whole boom is a less visible mechanism. Malta’s recent growth has leaned heavily on importing third-country nationals to do the work the Maltese increasingly will not, or are not equipped to, do — catering, construction, nursing, elderly care. Those workers then need somewhere to live, which drives demand for yet more buildings, which requires yet more imported labour to construct them. It is a cycle, and a self-referential one: growth that largely consists of adding people, housing the people, and counting the housing. As an engine of genuine, compounding prosperity, it is profoundly flawed, and laden with negative externalities.

Each of these moves is, in isolation, a defensible commercial decision. Taken together, over the two decades in which Taiwan was compounding its engineering lead, they amount — in my view — to a country choosing rent, and rent-seeking, as an industry and as a national strategy, over building of skill and intellectual property.

And the bill for that choice has begun to arrive.

The fork in the road — two trajectories, side by side

The same decades, two islands:

1949 — Taiwan — Retreat across the strait. The Republic of China government relocates to Taiwan: a poor, war-damaged agrarian society of a few million people, with per-capita income near $150 a year.
1950 — Malta — An island that is a naval base. Malta’s economy is, in effect, the British armed forces with an island attached. The Royal Naval Dockyard is the single largest employer. The work is genuinely skilled — but ship repair is a mobile, price-competitive trade, and over the coming decades it will drift to cheaper yards in Turkey, then Bangladesh and beyond.
1964 — Malta — Independence from Britain. With the British base winding down, the island faces an existential question: how does a place this small earn a living?
1966 — Taiwan — The Kaohsiung export zone. Taiwan opens the world’s first modern export-processing zone, putting cheap labour to work assembling textiles, plastics and toys — the “mass Barbie production” era.
1973 — Taiwan — ITRI, and a deal with RCA. Taiwan founds the Industrial Technology Research Institute. Three years later, in 1976, it negotiates a technology-transfer deal with America’s RCA — importing chip-making know-how and training a generation of engineers.
1974 — Malta — Playmobil arrives. The German toy-maker begins manufacturing in Malta. It still makes essentially every Playmobil figure in the world here — proof that skilled industry can root on the island.
1980s — Malta — The textile push. Malta makes a real bid in textiles, plastics and consumer goods. This was the fork in the road: a chance to treat low-cost assembly as a school, not a destination.
1987 — Taiwan — TSMC, and democracy. Morris Chang spins TSMC out of ITRI, with the Taiwanese state putting up roughly half the founding capital. The same year, Taiwan lifts martial law.
1990s — Malta — Electronics, then EU. Engineering and electronics firms arrive, anchored by ST Microelectronics. Malta sets its course toward EU membership and an open, services-led economy.
2004–2008 — Malta — Into the EU and the euro. EU accession and euro adoption open a faster road than industrial upgrading: regulatory arbitrage. The Baltic states join in the same 2004 enlargement; they will use the next two decades very differently.
2014 — Malta — Selling the passport. Malta launches its Individual Investor Programme, selling citizenship — and with it EU citizenship — to wealthy foreigners.
2018 — Malta — “Blockchain Island.” Malta passes a bespoke crypto law and courts the world’s exchanges. The branding outpaces the substance; reputational risk follows.
2021 — Malta — The grey list. The Financial Action Task Force places Malta on its money-laundering “grey list.” Malta exits after about a year, but the reputational damage lingers.
2020s — Taiwan — The silicon shield. TSMC now makes over 90% of the world’s most advanced chips. An AI-driven boom sends Taiwan’s growth to its fastest in decades.
2025 — Malta — The court closes the shop. The European Court of Justice rules Malta’s passport-sale scheme illegal, calling it the commodification of citizenship.

What a country spends on knowing more

If I were allowed only one statistic to predict which of two economies will still be prospering in 2050, I would pick this one: the share of national income spent on research and development. This figure is, in effect, the rate at which a country buys its own future.

Taiwan spends close to 4% of everything it earns on research and development — a figure that places it third on Earth, behind only Israel and South Korea. [25] Malta spends around 0.5%, which places it close to the bottom of the entire European Union, alongside Romania. [24] The Maltese government has stated an ambition to reach 2% by 2030; [41] for now, the gap is roughly eightfold.

This is not an abstract accounting difference. It is the compounding mechanism itself. A country spending 4% on R&D is buying patents, training PhD-level engineers, and seeding the firms that will be globally dominant in twenty years. A country spending 0.5% is, broadly, renting out what it already has — its location, its passport, its tax code, its coastline — and hoping the tenants do not leave.

Twenty years, not one unicorn

A “unicorn,” in the language of technology, is a privately held start-up that grows to a valuation above one billion dollars. It is a crude measure of a country’s capacity to grow something globally significant from nothing. Here is a comparison that should sting.

Malta joined the European Union in 2004. So, in the very same enlargement, did Estonia, Latvia and Lithuania — the three Baltic states. Comparably small. Comparably peripheral. Comparably starting their modern economic lives at the same moment, with the same access to the same single market.

Malta, since 2004: zero. Not one home-grown unicorn. The genuine bright spots are few and include: Hotjar (web analytics, sold to France’s Contentsquare), Altaro (backup software, sold to Hornetsecurity) and eCabs (home-grown ride-hailing).
Estonia, alone: around ten. Skype, Wise, Bolt, Pipedrive, Playtech and more — roughly ten unicorns, and about 7.7 per million inhabitants, the highest rate in Europe. Lithuania and Latvia have since minted their own.

Estonia, with a population only a little larger than Malta’s, has produced around ten billion-dollar companies and a string of platforms the world actually uses. Malta, in the same twenty years, has produced exactly none. [43]

This is not, I want to be clear, a story about a shortage of Maltese talent. Hotjar, Altaro and eCabs are real, creditable companies built by capable people. It is a story about soil — about whether a country builds the capital, the engineers, the institutions and the patient tolerance for a long, uncertain decade in which something enormous can take root. The Baltics built that soil. Malta, busy selling passports and licences, mostly did not.

Limestone, when we could have had silicon

Here I will stop reporting facts and figures, and start arguing. What follows is opinion. I believe Malta’s last twenty years represent a failure of political foresight and political will.

Start with the chemistry, because it turns out to be a better metaphor than I first realised. Malta has one significant natural mineral resource, and it is limestone — calcium carbonate, quarried, cut, and turned into the buildings of the islands. [35] A microchip, meanwhile, is built on silicon, and silicon comes from silica — silicon dioxide — which is simply quartz, the principal ingredient of ordinary sand.

Here is the elegant part. The carbon at the heart of limestone and the silicon at the heart of a chip are not strangers: they sit directly above and below one another in Group 14 of the periodic table — chemical siblings, both able to form four bonds, both among the great rock-building elements of the Earth’s crust. Limestone and silica are, in that sense, cousins: two of the cheapest, commonest, most abundant materials on the planet. You can pick either of them up off the ground for nothing.

So the difference between a Maltese apartment block and a Taiwanese microchip was never the mineral. It was the knowledge a nation chose to add to it.

Taiwan took ordinary sand and, over fifty years of schooling and institutes and patient failure, refined it to a purity of eleven nines and printed cities of logic onto it.

Limestone you cut with a saw and stack.

And here is the irony that completes the metaphor. Malta did not even hold on to the limestone. We over-extracted that too. Buildings of traditional globigerina limestone — franka — are now a rarity: the readily quarriable stone has been largely worked out, and the masons with the skill to cut and lay it have dwindled to a scarce few. For twenty years now, the island has built itself overwhelmingly in cheap concrete block rather than in any kind of vernacular stone. We exhausted the easy resource, let the craft that worked it decay, and replaced it with something cheaper and more generic still. Knowledge added, in one island. Knowledge lost, in the other.

Silicon is what you get when a country decides to become excellent, in fact, world-leading at something incredibly hard, and then refuses to be distracted for fifty years.

My argument against Maltese policy is not that it manufactured nothing; Playmobil and ST Microelectronics and a dozen other smaller, not-as-well-known firms prove that real, skilled, exportable industry could thrive on the island. My argument is that these were treated as happy accidents rather than as a strategy to be deepened. Instead of going all-in on a few highly specific, highly specialised niches — instead of finishing the climb the 1980s textile push had started — the national effort was spent, again and again, on whatever was quickest:

The passport shop. From 2014, Malta sold citizenship — and with it, EU citizenship — to wealthy foreigners. In April 2025 the European Court of Justice ruled the scheme illegal, calling it the commodification of nationality. [26]
“Blockchain Island.” The 2018 bid to be the world’s crypto capital generated headlines and a bespoke law. The substance was thinner; reputational risk and EU friction followed. [40]
Build, build, build. An economic model leaning on construction, property, and the imported labour that fills and rents the buildings — making overdevelopment the public’s single greatest stated concern. [38]
Footloose registries. Flags of convenience, ship and yacht registrations, online gambling licences: commoditised, mobile industries that will relocate the moment a more tax-efficient harbour opens. [28]

The through-line connecting these is fragility. A chip fab is, almost by design, immovable: it represents tens of billions of dollars of sunk capital and a workforce that took a generation to train. Consider just one link in that chain — ASML, the Dutch firm that is the sole maker on Earth of the extreme-ultraviolet machines a leading-edge fab cannot operate without. At the end of 2025 ASML sat on an order backlog of roughly €38.8 billion, with its machines booked solid through 2027. [44] That is what a genuinely defensible industry looks like: a queue years long, a near-monopoly, and customers who cannot simply shop elsewhere. It is immovable, non-transferable — and extraordinarily lucrative precisely because of that.

A gambling licence-holder, a crypto exchange or a registered yacht is the opposite. It has no roots. It is in Malta because of a spreadsheet, and it will leave the day the spreadsheet changes — when another jurisdiction undercuts the tax rate, or when Brussels, as it increasingly does, decides the arbitrage must end.

There were signs of the bill arriving even before the passport ruling. In 2021 the Financial Action Task Force placed Malta — uniquely among EU states at the time — on its money-laundering “grey list,” a direct consequence of the reputation a sell-access economy accumulates. [42] Malta worked its way off the list within about a year, which is to its credit. But the reputational damage was done, and anecdotally, we have no idea how many billion-euro opportunities passed us by simply because of it — because somewhere, in some boardroom, the conversation went: “What, Malta? Grey-listed last year? Where they blew up that journalist? No.” And grey-listing is the kind of event that simply does not happen to a country whose primary export is precision-engineered components.

A semiconductor fab cannot pack up and leave. A passport buyer, a crypto exchange and a flag of convenience can — and one day will.

In fairness — the case against my own argument

An honest essay has to argue against itself, so here are the strongest objections to everything above.

Malta is genuinely wealthy, and that is not nothing. On GDP per head, adjusted for the cost of living, Malta sits comfortably among rich nations. Unemployment is low. The services-and-arbitrage model has delivered two decades of real growth and real jobs. A Maltese reader is entitled to ask: if the strategy was so foolish, why are we this prosperous? “Il-Malti sinjur żgħir sar, u kullħadd sid ta’ daru!” — the Maltese have become comfortably well-off, the saying runs, and everyone the owner of their own home. It is a fair question, and the honest answer is that the critique here is about foundations and resilience, not present comfort. Also note that I am not discussing quality of life here at all, although it is known to be an increasingly grave concern for many Maltese; however, since I did not present any data — quantitative, anecdotal, or otherwise — about Taiwan, I feel that would make any comparison unfair.

And yet I fear that the boom years — the past twenty — will be followed by a bust, caused by a plethora of factors: a lack of forethought and resilience that was never built into our model; rising fuel prices; an increasingly complex and dangerous geopolitical situation; a total lack of food-production sovereignty and food-supply security; the over-extraction of water; dependency on our neighbours for electrical power; and the basic exposure of being a small island entirely at the mercy of the interplay of global politics and global markets. We have hedged against none of the above — choosing instead to blindly build apartments, buy cars, and keep our money in savings accounts, rather than building out new industrial bases and real financial might.

The scale and geography are not the same. Taiwan has 23 million people; Malta has barely half a million. A semiconductor industry needs a deep pool of engineers, a domestic supply chain and an enormous capital base — things a micro-state may simply be unable to assemble. My argument is not that Malta could have literally built TSMC. My argument is that Malta could have built literally almost anything else, but instead it gave out gambling licences, sold passports, and built flats.

Malta has, in one crucial respect, been far luckier than Taiwan. And this genuinely cuts against me. Taiwan built its miracle under permanent existential threat — a giant, often belligerent neighbour ninety miles to the north that has never renounced the use of force against it. Malta’s neighbour to the north is the European continent: large, fragmented, perhaps indifferent beyond what the EU charter obliges it to be — but not a hostile power, not a standing military menace. That safety is a real and underrated blessing. But it is also double-edged. The very threat that endangered Taiwan also disciplined it, forging an unusual national unity behind long-term strategy. Malta’s safety removed that forcing function — and made the easy roads available.

Specialisation is also a risk. Taiwan’s concentration in semiconductors is a strength, but it is also a single point of failure: a blockade, an earthquake or a sharp turn in the chip cycle would expose the island brutally. Malta’s diversified, low-intensity portfolio is, in one sense, hedged. There is a respectable argument that a tiny state should stay nimble rather than bet the island on one industry.

I accept all of that. My argument is narrower, and I think harder to dismiss: that between roughly 2004 and 2024, Malta repeatedly chose the quick rent over the slow skill — and that it spent on selling access the political energy, the institutional attention and the patient capital that a handful of deep, defensible, genuinely Maltese specialisms would have required. Taiwan is not a blueprint to be copied. It is a mirror held up to a question of will.

What a surplus is for

There is one more idea from the Search Engine episode that has stayed with me, and it widens the question beyond chips and GDP.

A society with a deep, genuinely productive industrial base can afford things that do not immediately pay. It can afford to support its people in the humanities, in the liberal arts, in scholarship and music and the fine arts — because the wealth-generating engine underneath is real, it compounds, and it throws off a surplus large enough to underwrite softer, more human kinds of development. A full, dignified life lived within the arts becomes genuinely possible when the economy beneath it is rich and self-sustaining.

Malta is the counter-case. Because the model is extractive rather than generative — rent rather than production — there is no comparable surplus left over. There is no broad industrial base. There is no sovereign wealth fund of the kind the oil states built, no diversified manufacturing engine of the kind Taiwan has. And so, on the island, a full-time living made purely within the arts or the humanities is, for most people, barely feasible — not for any want of Maltese talent or feeling, but because the economic foundation that would have underwritten it was never built. A productive base buys a society the freedom to value things that do not turn an immediate profit. An extractive one keeps everyone hustling.

The roads not taken

Shelley Rigger’s Taiwan is, above all, a story about agency — about a place that was small, threatened and poor, and that decided its smallness would not be its destiny. It chose a hard road and walked it for half a century without flinching.

Malta is also small, and also resourceful, and also full of capable people.

The Playmobil factory, the ST Microelectronics plant, eCabs, Hotjar, and a few dozen other firms of all kinds of sizes — but, crucially, of deep technical specialisation — are proof that the harder road was always open to us.

We mostly did not take it. We preferred the beaten path, the easier to navigate, more obvious, quicker route to returns.

Whether our short-term thinking will continue to pay dividends remains to be seen.

Whether the next twenty years rhyme with the last twenty is not written in the limestone — so let the chips fall where they may.

There is though, as there always was, a choice.

Sources & method

Every quantitative claim above is referenced below, and double-sourced wherever a second independent body publishes the same measure. Bodies cited include the IMF, the World Bank, Eurostat, the Asian Development Bank, the Taiwan and European patent offices, national statistics offices, the World Happiness Report and the European Court of Justice. Interpretation and opinion are the author’s own.

The episodes that started this. “The Many Lives of Taiwan,” from the podcast Search Engine, hosted by PJ Vogt — the formative source for this essay’s framing of Taiwan’s arc from colonial subjugation, through mass toy production, to technological powerhouse, and for the idea that a productive base is what lets a society afford the humanities. The episode’s guest is Professor Shelley Rigger of Davidson College, whose books Why Taiwan Matters: Small Island, Global Powerhouse and The Tiger Leading the Dragon inform the historical account here. The detail of the Taiwanese state’s return on its TSMC investment draws on the Acquired podcast’s history of TSMC, corroborated with shareholding data. Any errors of compression or emphasis are mine, not the podcasters’.

Shelley Rigger — biography and works, Davidson College / UC Berkeley Institute of East Asian Studies.
Search Engine — “The Many Lives of Taiwan,” episode description and audio (host PJ Vogt; guest Shelley Rigger).
Acquired podcast — “TSMC” episode (acquired.fm), on the Taiwanese government’s founding stake.
CIA World Factbook / Wikipedia, “Geography of Taiwan” and Yale EPI country profile — Taiwan land area ≈ 36,200 km².
Eurydice / European Commission — Malta land area, just over 316 km².
Wikipedia, “Economy of Taiwan” (citing the IMF), with Worldometer and the Asian Development Bank — Taiwan population ≈ 23.2 million.
NSO Malta (World Population Day 2025) and Malta Today — Malta population ≈ 574,250 at end-2024.
NSO Malta and Newsbook — natural increase fell from 432 (2023) to 193 (2024); growth “mainly dependent” on net migration. The Shift News, analysing NSO data — Maltese-family births added ≈ 5,660 over 2014–24 while foreign residents grew by ≈ 131,000.
NSO Malta / Georank (Eurostat-based data) and Wikipedia (CIA data) — Malta total fertility rate ≈ 1.1–1.5, below the 2.1 replacement level since ≈ 1990; Finance Minister Clyde Caruana quoted via The Shift News.
Asian Development Bank, Key Indicators Database — Taiwan population change ≈ 0.2% (2024); both islands have sub-replacement fertility.
Wikipedia, “Economy of Taiwan” (IMF) and Statista (IMF series) — Taiwan nominal GDP ≈ $884 bn (2025).
Wikipedia, “Economy of Taiwan” (IMF) and Statista (IMF series) — Taiwan nominal GDP ≈ $884 bn (2025).
Wikipedia, “Economy of Malta” (IMF) and Trading Economics — Malta nominal GDP ≈ $22.7–24.3 bn (2024).
Wikipedia, “Economy of Malta” citing the IMF — Malta GDP per capita (nominal) ≈ $41,700 (2024).
Taiwan Government Portal (IMF, April 2025) and Statista (IMF, December 2025) — Taiwan GDP per capita ≈ $34,430 rising toward ≈ $39,490.
Wikipedia / IMF and World Bank / Trading Economics — Malta GDP per capita PPP ≈ $60,500–67,700 (2024).
Wikipedia / IMF and the Taiwan Government Portal — Taiwan GDP per capita PPP, $76k–$98k range (2024–26 estimates).
Wikipedia, “Economy of Malta” and WorldAtlas — manufacturing ≈ 10–11% of GDP; services ≈ 87%.
Wikipedia, “Economy of Taiwan” — industry ≈ 36% of GDP.
Eurostat (via Wikipedia, “Economy of Malta”) and the Malta Ministry for Finance, Economic Survey 2024 — Gini ≈ 28–31, trending upward.
Wikipedia, “Economy of Taiwan” and World Economics — Taiwan Gini ≈ 33.9 (2023 est.). Methodologies differ from Eurostat’s; compare loosely.
World Happiness Report (2024 edition) and Times of Malta — Malta ≈ 40th, score ≈ 6.35; Maltese under-30s ranked the unhappiest youth in the EU.
World Happiness Report 2025 and the Global Taiwan Institute / Time Out — Taiwan 27th globally, the happiest place in East Asia.
Eurostat, R&D expenditure statistics and Malta NSO / Business Now — Malta R&D ≈ 0.5–0.6% of GDP, among the EU’s lowest.
UK Science & Innovation Network — Taiwan factsheet — Taiwan R&D ≈ 3.96% of GDP (2022), third globally behind Israel and South Korea.
Reuters and the ECPR’s The Loop / ASIL / eucrim — European Court of Justice ruling of 29 April 2025 (Commission v Malta, C-181/23); the scheme raised > €1.4 bn since 2015.
Economics Observatory and The Conversation — TSMC makes >90% of the world’s most advanced chips; ≈ 8% of Taiwan’s GDP and 12% of exports; the “silicon shield.”
Fitch Ratings (via SBC News) and VIXIO regulatory analysis — online gambling ≈ 12% of Malta’s GDP.
taiwan.md (“Economic Miracle” / “Industrial Transformation”) — Taiwan’s 1951 GDP per capita ≈ US$154; the 1950s land reforms.
taiwan.md and Asterisk Magazine — the Kaohsiung export-processing zone (1966); ≈ 1.7 million SMEs underpinning the economy.
Wikipedia, “Semiconductor industry in Taiwan” and The Conversation — ITRI founded 1973; the 1976 RCA technology-transfer project; K. T. Li.
Wikipedia, “TSMC” and Domino Theory — TSMC founded 1987; the Taiwanese state provided roughly half the founding capital.
Kamil Franek Business Analytics, Domino Theory and CB Insights — the National Development Fund retains ≈ 6.4% of TSMC, a stake worth ≈ US$24.6 bn at end-2022 and more since.
Taiwan Intellectual Property Office (TIPO) annual reports — ≈ 50,000 invention-patent applications filed in Taiwan each year (2022–24). Statista / European Patent Office — Malta ≈ 140 EPO applications a year (2017 peak). The two offices are not identical; the per-head comparison is indicative.
Encyclopædia Britannica, “Malta — Economy” and the USGS Minerals Yearbook — the British-base legacy and dockyard; limestone as Malta’s only exploited mineral resource.
Alastair Philip Wiper / Playmobil reporting and The Worldfolio (Malta Industrial Parks) — Playmobil manufacturing in Malta since the mid-1970s.
The Worldfolio (Malta Industrial Parks) and WorldAtlas — ST Microelectronics among Malta’s largest manufacturers and exporters.
Wikipedia, “Immigration to Malta” (NSO figures) and MaltaToday surveys — foreign residents rose from ≈ 3% (2005) to ≈ 31% (2024); overpopulation now the top public concern.
Finance Magnates and Nomad Gate — Malta’s Individual Investor Programme (passport sales) launched 2014.
Finance Magnates, Bloomberg and CryptoTimes — Malta’s 2018 “Blockchain Island” branding and Virtual Financial Assets Act, and the subsequent EU friction.
IMF, Malta 2024 Article IV Consultation — R&D at 0.7% of GDP (2022); a government target of 2% by 2030.
Reuters and VIXIO / the Malta Employers’ Association — the Financial Action Task Force placed Malta on its “grey list” in June 2021; Malta exited after roughly a year.
Invest in Estonia, the Institute of Central Europe and Sifted — Estonia’s ≈ 10 unicorns (≈ 7.7 per million inhabitants); the Baltic states joined the EU in 2004 alongside Malta.
ASML Q4 2025 results, reported via MarketBeat and TradingKey — ASML, the sole maker of EUV lithography machines, ended 2025 with a ≈ €38.8 bn order backlog, with capacity booked through 2027.
QCAdvisor (2025) and the Reshoring Institute via Logistics Management / Modern Materials Handling — coastal Chinese manufacturing wages grew ≈ 9% a year over 2015–2024 to ≈ $6.50 an hour, several times Vietnam’s ≈ $2–3 and Bangladesh’s sub-$1; labour-intensive production is steadily relocating away from China.

About the author

Mark Debono is the founder of Systemato and an independent marketing and communications consultant. He has strong opinions about AI, has no problem talking about his failures, and is always looking for interesting companies who need a senior communications advisor.

Limestone & Silicon – what Taiwan did that Malta didn’t