A tech founder's economic argument for why Southeast Asia's largest economy cannot make its innovation flywheel spin. From rent-seeking capital allocation and a colonial trust deficit to missing exit markets, every stage of the cycle is broken.
Indonesia has 280 million people, a perfect climate, extraordinary natural resources, and a generation of ambitious founders. It also has an innovation ecosystem that runs in place. This is an attempt to explain why, using economics rather than anecdotes.
In a functioning innovation ecosystem, capital flows in a circle. The circle has six stages, and every stage must work for the system to compound.
Venture capital or government grants fund startups. Those startups burn money competing for customers in open markets. The best ones, selected by merit, find product-market fit. The companies that reach scale exit through IPO or acquisition by private equity. Exit proceeds return to investors and founders, who deploy them into the next generation of startups. The cycle repeats.
This is not a Silicon Valley myth. It is a well-documented economic phenomenon that follows the same logic as Keynes’s circular flow of income, but with a critical amplifier: each revolution produces not just financial returns but also knowledge, networks, and institutional trust that make the next revolution more efficient.
The total value created by $n$ revolutions of the flywheel can be modeled as a compounding series:
\[V_n = C_0 \cdot \prod_{t=1}^{n} \bigl(1 + r_t \cdot \eta_t\bigr)\]where $C_0$ is the initial capital entering the ecosystem, $r_t$ is the return on innovation in revolution $t$, and $\eta_t$ is the recycling efficiency: the fraction of exit proceeds that re-enter the ecosystem as new venture capital. When $\eta_t$ is high, this is compound growth. When $\eta_t$ is near zero, the product collapses to a single term and every generation of startups starts from scratch.
In the United States, $\eta_t$ approaches 0.7 to 0.8. Most exit proceeds flow back: investors raise new funds, founders become angels, equity-compensated employees start their own companies. In Indonesia, $\eta_t$ is close to zero.
The figure below shows the healthy flywheel. Every arrow represents a flow of capital and knowledge. Every node is a stage that must be meritocratic for the system to work.
The flywheel is powerful precisely because it is circular. Remove any single stage, and the system degrades to a linear, one-shot model where every dollar of innovation funding produces returns once and then disappears. The question for Indonesia is not whether it has smart people or big markets. It does. The question is whether the flywheel can spin.
It cannot.
Indonesia’s innovation cycle fails at three stages simultaneously. Each failure reinforces the others, creating a system that does not merely slow down but locks in place.
The failure is systemic: without exits, investors lose money, making them less willing to deploy capital. Without capital, startups cannot compete. Without competition, no one reaches the scale needed for an exit. Each break reinforces the others.
We can formalize this as a system of coupled dependencies. Let $C$ be available capital, $S$ be startup quality, and $E$ be exit availability. In a healthy ecosystem:
\[C_{t+1} = f(E_t), \quad S_{t+1} = g(C_t), \quad E_{t+1} = h(S_t)\]Each variable depends on the previous period’s value of another. If any one drops to zero, all three converge to zero within a few periods. Indonesia is stuck at approximately $(C, S, E) \to (0, 0, 0)$, and escaping this fixed point requires a simultaneous shock to all three variables.
Indonesia’s government innovation grants suffer from a textbook principal-agent problem. The government (principal) allocates funds intended for innovation, but the bureaucrats administering those funds (agents) have misaligned incentives.
In Tullock’s rent-seeking framework, the expected payoff from lobbying for a government grant is:
\[E[\pi_i] = \frac{x_i}{\sum_{j=1}^{n} x_j} \cdot R - x_i\]where $x_i$ is the resources spent on lobbying by firm $i$, $R$ is the total grant value, and $n$ is the number of competitors. In a meritocratic system, $x_i$ represents genuine innovation effort: building a better product, publishing stronger research, demonstrating clearer market traction. The firms that invest the most in quality win.
In Indonesia, $x_i$ represents something else entirely: connections, relationship maintenance, and the understanding that grant money will be “shared equally among government stakeholders.” The Nash equilibrium shifts from quality competition to patronage competition, and the total social waste, what Tullock called the “welfare cost of rent-seeking,” equals the sum of all lobbying expenditures:
\[W = \sum_{i=1}^{n} x_i^{*}\]This waste is not just the money spent on lobbying itself. It is also the opportunity cost: every hour a founder spends cultivating bureaucratic connections is an hour not spent building a product. The deadweight loss extends far beyond the direct cost.
The practical result: the flow of money moving downward from government to actual innovation is always much smaller than the headline number suggests. A 100 billion rupiah innovation grant might produce 20 billion rupiah worth of actual research, with the rest absorbed by the rent-seeking apparatus. By the time capital reaches a lab or a startup, it has lost most of its force.
Every country has corruption. What makes Indonesia’s version particularly destructive for innovation is that it operates at the capital-entry stage, the very first node of the flywheel. When corruption happens at later stages (say, a company bribing officials for permits), the innovation has already occurred. When it happens at the entry stage, the innovation never starts.
Krueger’s 1974 formalization captures this. She showed that when government controls access to economic rents, the resources devoted to capturing those rents can equal or exceed the rents themselves:
\[\text{DWL}_{rent\text{-}seeking} \geq R\]The deadweight loss from rent-seeking can exceed the total value of the prize being fought over. In Indonesia’s case, the grant system may actually destroy more innovative capacity than it creates.
The venture capital ecosystem in Indonesia suffers from Akerlof’s information asymmetry problem. In his 1970 paper The Market for “Lemons”: Quality Uncertainty and the Market Mechanism (Quarterly Journal of Economics, awarded the Nobel Prize in 2001), Akerlof showed that when buyers cannot distinguish high-quality goods from junk, the market collapses.
Applied to the Indonesian VC ecosystem: the expected value of any individual startup deal is
\[E[V] = q \cdot V_H + (1-q) \cdot V_L\]where $q$ is the probability that a startup is genuinely high-quality, $V_H$ is the value of a good startup, and $V_L$ is the value of a lemon. When the ecosystem is flooded with dishonest actors, founders who overpromise, fabricate metrics, and misrepresent traction, $q$ drops. And the expected value of any individual deal drops with it.
Rational VCs respond in one of two ways: withdraw from the market entirely, or demand increasingly punitive deal terms (higher equity stakes, more control provisions, shorter runways) to compensate for the elevated risk of lemons. Both responses make things worse.
Punitive terms drive away the honest founders who have alternatives. A talented Indonesian engineer with a strong product can raise money in Singapore, where the VC ecosystem is more mature and deal terms are more founder-friendly. So the best founders leave, which further reduces $q$ in the domestic market, which further tightens terms, which drives away more good founders. This is the classic Akerlof death spiral: the market converges on an equilibrium dominated by the lowest-quality participants.
In equilibrium, the willingness-to-pay of VCs converges to:
\[\text{WTP}^{*} = V_L + \varepsilon\]Investors will only pay slightly above the lemon value, which is insufficient to fund genuine innovation. The market clears at a price that only lemons will accept.
Indonesia’s trust deficit has deep roots that neither the government nor the startup ecosystem created. For centuries, business in the archipelago ran on oral promises: a handshake, a community reputation, a family name. This worked in small, high-trust networks where everyone knew everyone and social enforcement was immediate. Default on a promise, and the entire village knew by sundown.
Then the Dutch arrived and imposed a legal system that demanded written proof for everything. Contracts, notarized documents, stamped approvals. The formal institution was transplanted wholesale onto a society whose informal institutions operated on completely different principles.
Douglass North’s institutional economics framework, developed in Institutions, Institutional Change and Economic Performance (1990, awarded the Nobel Prize in 1993), explains why this matters. North decomposed transaction costs into three categories:
\[TC = TC_{search} + TC_{negotiation} + TC_{enforcement}\]Total transaction costs ($TC$) are the sum of search costs (finding trustworthy partners), negotiation costs (agreeing on terms and committing them to paper), and enforcement costs (ensuring compliance when disputes arise).
When formal institutions (written law, contracts, courts) and informal institutions (social norms, oral trust, community reputation) are aligned, these costs are low. People trust contracts because the courts enforce them, and people trust each other because social norms reinforce honesty. The two systems are complementary.
When they conflict, as they do in Indonesia, every transaction bears an institutional tax.
The practical effect: every business deal in Indonesia requires more paperwork, more lawyers, more time, and more suspicion than equivalent deals in ecosystems where institutions are internally consistent. An Indonesian oral promise carries real social weight, but no legal standing. A written contract carries legal standing, but culturally feels like an insult, an implication that your word alone is not enough.
This institutional friction compounds at every stage of the innovation flywheel. VCs spend disproportionate time on due diligence. Startups burn months on legal compliance that their Singaporean competitors complete in days. Partnerships form slowly and dissolve acrimoniously because neither the written nor the oral enforcement mechanism works reliably.
The velocity of the flywheel, how fast capital and ideas can circulate, is inversely proportional to transaction costs:
\[v = \frac{k}{TC}\]where $k$ is a constant representing the intrinsic quality of opportunities in the market and $TC$ is the total transaction cost per deal. Indonesia’s high $TC$ means the flywheel turns slowly even when $k$ is high, even when the ideas and the market opportunity are genuinely good.
Indonesia sits on some of the most fertile land on Earth, blessed with a tropical climate that produces year-round harvests, rich fisheries, and extraordinary biodiversity. For most of human history, this was an unambiguous advantage. But in the context of building an innovation-driven economy, it creates what economists call the resource curse.
The Solow growth model, introduced in Robert Solow’s A Contribution to the Theory of Economic Growth (Quarterly Journal of Economics, 1956; Nobel Prize 1987), decomposes economic output as:
\[Y = A \cdot K^{\alpha} \cdot L^{1-\alpha}\]where $Y$ is total output, $A$ is total factor productivity (TFP, the “innovation term”), $K$ is physical capital, $L$ is labor, and $\alpha \in (0,1)$ is the capital share of output. The critical variable for long-run growth is $A$: countries can accumulate more capital and labor, but only increases in $A$ generate sustained growth beyond the steady state.
Countries rich in natural resources can sustain a high $Y$ through large $K$ (extracting and selling resources) and abundant $L$ (agriculture in a fertile climate) without ever significantly improving $A$. The economic pressure to innovate, to push TFP higher, simply does not exist at the intensity it exists in resource-scarce economies.
Sachs and Warner formalized this as the “resource curse” in their 1995 NBER working paper Natural Resource Abundance and Economic Growth. They found a robust negative correlation between natural resource dependence and economic growth across 95 countries, controlling for other variables. The proposed mechanism: resource abundance shifts the economy toward extraction and rent-seeking and away from innovation and manufacturing, precisely because extraction is the easier path.
This is not a moral judgment about work ethic. It is a structural observation about incentive landscapes. Our ancestors in the archipelago did not need the obsessive long-term planning, the tolerance for repeated failure, or the relentless optimization that characterize innovation cultures, because the environment was generous enough that these adaptations were not survival-critical. Rice grows three times a year. Fish are abundant. The climate does not kill you if you fail to prepare.
The cognitive and cultural habits that drive innovation ecosystems, delayed gratification, comfort with ambiguity, obsessive iteration, these are not innate traits. They are adaptations to scarcity, forged over generations in environments where failure to innovate meant death. Israel’s desert, South Korea’s post-war devastation, Singapore’s complete absence of natural resources: these constraints produced innovation cultures not because those populations are inherently superior, but because they had no alternative.
The implication for the innovation flywheel: the human capital feeding into Indonesian startups, while improving rapidly in Jakarta, Bandung, and Surabaya, operates against a cultural baseline calibrated for abundance rather than scarcity. The founder intensity required to push a startup through the valley of death is harder to sustain in an environment where the valley of death is, historically, not very deadly.
We can express this as a modification to the Solow model. Let $\phi$ be a “necessity multiplier” on TFP growth:
\[\dot{A} = \phi(\text{scarcity}) \cdot s \cdot Y - \delta \cdot A\]where $s$ is the savings/investment rate, $\delta$ is depreciation, and $\phi$ is increasing in scarcity. When resources are abundant, $\phi$ is small, and TFP grows slowly regardless of investment. Indonesia’s natural endowment, paradoxically, dampens the one variable that matters most for long-run growth.
Even if an Indonesian startup survives the capital problem, navigates the institutional friction, and builds a genuine product, it faces the final and most devastating break in the flywheel: there is nowhere to go.
In a functioning innovation ecosystem, exits serve two purposes. They reward risk-taking, which incentivizes future capital deployment. And they recycle capital, which fuels the next generation of startups. Both purposes require liquid, deep exit markets.
In the United States, a successful startup can exit through acquisition by a larger technology company (Google, Microsoft, Apple routinely acquire startups for talent and technology), private equity buyout (which provides liquidity to early investors while the company continues to grow), or IPO on a deep, liquid stock market with technology-literate investors (NASDAQ, NYSE).
In Indonesia, the options are: IPO on the IDX. That is essentially it.
The GoTo Group, formed by the 2021 merger of Gojek and Tokopedia, provides the clearest illustration. GoTo’s IPO in April 2022 was the largest in Indonesian history, raising $1.1 billion. By standard flywheel logic, this should have been a catalyst: early investors get returns, founders get liquidity, the success story attracts the next generation of founders and investors.
Instead, the capital went primarily to foreign investors (SoftBank Vision Fund, Alibaba, Google, Temasek) who deployed their returns into other markets, not back into Indonesian startups. Indonesian retail investors who bought at IPO watched the stock decline. The capital escaped the ecosystem entirely.
The recycling efficiency tells the story:
\[\eta_{US} = \frac{C_{recycled}}{C_{exit}} \approx 0.7 \text{ to } 0.8\] \[\eta_{ID} = \frac{C_{recycled}}{C_{exit}} \approx 0.05 \text{ to } 0.1\]When a US startup exits, most of the proceeds flow back into the venture ecosystem. Investors deploy into new funds. Founders become angel investors. Employees with equity start their own companies. The knowledge, networks, and capital circulate.
When an Indonesian startup IPOs on the IDX, the money goes to public market investors who have no connection to the innovation ecosystem, and to foreign institutional investors who redeploy the capital to Singapore, Beijing, or the Bay Area. The capital escapes, and the flywheel loses its momentum.
Without a functioning exit market, the entire venture capital model breaks down. VC is a power-law business: a few massive exits pay for all the failures. If those exits cannot happen, or if they happen but the capital does not recycle, rational investors will not deploy capital into the ecosystem.
Schumpeter called this “creative destruction”: old firms are destroyed by innovative new ones, and the capital released by the destruction funds the next wave of creation. In Indonesia, the destruction happens (old firms lose share to Gojek, Tokopedia, Bukalapak), but the creation part of the cycle is severed. Capital is destroyed, not recycled.
The economic consequence of a broken innovation flywheel is devastating when viewed through the lens of compound growth.
Recall the compounding formula:
\[V_n = C_0 \cdot \prod_{t=1}^{n} \bigl(1 + r_t \cdot \eta_t\bigr)\]When $\eta_t \approx 0.75$ (US), and $r_t \approx 3$ (a reasonable venture return multiple), the growth factor per revolution is $1 + 3 \times 0.75 = 3.25$. After five revolutions:
\[V_5 = C_0 \cdot 3.25^5 \approx 3{,}500 \cdot C_0\]The initial capital has been multiplied 3,500 times. This is how Silicon Valley grew from a few semiconductor companies in the 1960s to a multi-trillion-dollar ecosystem.
When $\eta_t \approx 0.05$ (Indonesia), the growth factor per revolution is $1 + 3 \times 0.05 = 1.15$. After five revolutions:
\[V_5 = C_0 \cdot 1.15^5 \approx 2 \cdot C_0\]The capital has barely doubled. And that is the optimistic scenario, the one where five full revolutions actually occur. In practice, with a broken exit market and poisoned capital entry, Indonesia may never complete even one full revolution.
The divergence between these two paths is not linear. It is exponential. Every year that the flywheel stays locked, the gap between Indonesia and functioning innovation ecosystems widens, not by a fixed amount, but by a growing multiple.
This is why Indonesia’s tech scene feels like Groundhog Day. The same problems get solved by the same type of founders, who raise money from the same small pool of investors, build companies that reach the same ceiling, and then: nothing. The knowledge disperses. The capital evaporates. The next generation starts from zero.
There is no institutional memory, because there is no institution. There is no recycled capital, because there is no exit. There is no experienced founder network, because the experienced founders moved to Singapore.
I do not have a clean prescription. If I did, I would be executing it instead of writing about it. But the economic framework points to leverage points, and honest analysis is worth more than false optimism.
Fix the exit problem first. Everything else is downstream. The flywheel cannot spin without capital recycling. This means creating mechanisms for secondary sales of startup equity, encouraging regional PE funds that specialize in Southeast Asian technology companies, building acquisition pathways through ASEAN-wide consolidation, or creating IDX listing rules that are more friendly to technology companies with long revenue horizons.
The binding constraint is that Indonesian public markets are dominated by banking, mining, and consumer goods conglomerates. Technology companies with negative margins and high growth rates do not fit the IDX investor base. A NASDAQ-equivalent for Southeast Asian technology, possibly based in Jakarta, possibly a joint ASEAN initiative, would be transformative.
Reduce transaction costs through digital trust. Indonesia’s digital identity and fintech infrastructure (GoPay, OVO, DANA, Bank Jago) is actually ahead of many developed countries. Extend that logic from consumer payments to business formation, contract enforcement, and dispute resolution. If two parties can transact on GoPay with near-zero friction, the same infrastructure could underpin startup incorporation, cap table management, and investor agreements.
The institutional mismatch between oral trust and colonial paper law will not be resolved by cultural change, that takes generations, but it can be circumvented by technology that makes the paper layer fast, cheap, and invisible.
Import the flywheel temporarily. Singapore and the United States already have functioning innovation flywheels. Rather than building one from scratch, create regulatory and tax incentives for foreign VCs and PE firms to operate in Indonesia with exit pathways that connect to deeper capital markets. A Singaporean VC that can invest in an Indonesian startup, list it on SGX, and recycle the proceeds into the next Indonesian deal: that is a functioning flywheel, even if the recycling happens through Singapore.
This is not a permanent solution. Dependency on foreign capital markets perpetuates the extractive dynamic. But it is a way to start the flywheel spinning while domestic institutions catch up.
Accept that the climate comfort trap is real and work with it, not against it. Indonesia will not become South Korea. The cultural baseline is different, and pretending otherwise is a recipe for frustration. Instead, build innovation incentives that work with Indonesian strengths: the enormous domestic market (280 million consumers), the young and increasingly digital population, the geographic position as the gateway between the Indian and Pacific Oceans. The pressure to innovate does not need to come from existential scarcity. It can come from the sheer scale of opportunity, if the flywheel is structured to capture it.
The alternative is what we have now: a country with the fourth-largest population on Earth, extraordinary natural resources, a generation of founders who are as talented as their counterparts anywhere in the world, and an innovation ecosystem that cannot turn its potential into compound growth. The flywheel has all the pieces. They are just not connected.