As the coffers fill, more spoons reach in. With next year's national tax revenue set to rise sharply on the semiconductor boom, the government has expanded total spending to 820.9 trillion won. The 12.8% growth rate is the highest on record. The problem is that semiconductors are a classic cyclical industry. What happens if the government treats boom-time revenue as normal revenue, ramps up spending and then the chip cycle turns? Once expanded, spending is hard to unwind. In the National Assembly's budget review, the first question should be whether boom-dependent spending increases are sustainable, rather than line-by-line changes in individual programs. In an interview with Seoul Economic Daily on the 21st, Kim Woo-chul, a professor of taxation at the University of Seoul and president of the Korean Association of Public Finance, said the government must distinguish between revenue generated by rising chip demand and revenue generated by surging prices. "The budget should be built on structural revenue that strips out cyclical swings," he said.
Q: Is a 12.8% increase in government spending next year justified?
A: It is excessive. Seven to eight percent would be enough. Even in past periods of aggressive fiscal policy, spending growth was around 8-9%, and during COVID-19 it did not exceed 10%. The scale of the budget itself is now far larger than then, so even a 7-8% increase amounts to substantial fiscal expansion. The premise that potential growth rises if you spend 12.8% but not if you spend 7-8% is wrong from the start. Moreover, nominal growth and tax revenue are both rising quickly at the moment. Expanding fiscal policy sharply in that environment risks overstimulating aggregate demand and fueling instability in prices and asset markets.
Q: How should this revenue windfall have been used?
A: It should have been used to run a fiscal surplus and reduce national debt. Had spending growth been set at around 7-8%, roughly 36 trillion won less would have been spent and the managed fiscal balance could have posted a surplus in the 30 trillion won range. The increase of around 160 trillion won in tax revenue was a genuinely welcome opportunity for Korea's public finances. If the chip boom had continued, there was room for another surplus the following year. Two consecutive years of surplus in the managed fiscal balance could have changed the trajectory of Korean public finance itself. It is regrettable that the opportunity was missed.
Q: The government argues that investing now is what generates growth.
A: I am not denying that logic. The problem is the approach that assumes potential growth rises simply because you spend more. We have already run expansionary fiscal policy for years while accepting large deficits. In the past the banners were social investment, education and research and development; now they are artificial intelligence and semiconductors. The talk of "priming the pump" is the same. Did potential growth rise as a result? Japan also poured money into fiscal spending for a long period without lifting its potential growth rate.
Q: What would actually raise potential growth?
A: Regulatory reform and patient capital. Funds should be supplied long-term at low rates to fields the private sector finds too risky to enter readily. The government shares the risk in advanced sectors where success is uncertain. There is no need for the government to hand more cash to Samsung Electronics or SK hynix, which are already cash-rich. Companies competing directly in global markets have far more information and experience than the government does. It is difficult for the government to drive growth by picking promising industries and distributing subsidies. What is needed is to remove regulations, share risks the private sector cannot bear alone, and supply capital that is willing to wait a long time.
Q: Is the revenue increase from the chip boom sustainable?
A: Price and volume have to be separated. There is a possibility that chip demand, meaning volume, will rise structurally as AI spreads. But the current surge in corporate profits and tax revenue also owes a great deal to price effects. Prices move in cycles. Chinese producers are entering the market, and Samsung Electronics, SK hynix and Micron are all expanding capacity. As supply increases, the price effect weakens. Structural revenue arising from higher volume must be distinguished from temporary revenue arising from price spikes. Treating boom-time revenue as money that will keep coming in and raising spending accordingly is dangerous.
Q: How, then, do you identify "normal revenue"?
A: It is straightforward. If 100 should normally come in as taxes and 120 comes in, the additional revenue is 20. That means you first have to establish the baseline of 100. But the current practice takes the previous year, when revenue was weak, as the starting point, applies the average growth rate of the past decade, and calls anything above that additional revenue. The baseline itself is set incorrectly. You should use the potential growth rate to derive a normal level of gross domestic product and apply revenue elasticity. Structural revenue that strips out cyclical swings must be calculated first, and only money above that figure should be counted as additional revenue. Public spending should be set on the basis of normal revenue.
Q: What is the biggest problem with the Future Response Fund?
A: The fund is basically a vault. It locks money away to pass on to future generations. The sovereign wealth funds of Norway and oil-producing countries also began by accumulating surplus resources. In our case, however, revenue is being spent as revenue while national debt is increased to secure more than 100 trillion won in reserves. On the national balance sheet, that amounts to borrowing to fill a vault. That is why I have called it the world's first future fund built on debt. When the principal itself is debt, it is questionable whether withdrawing that money when needed is consistent with the fund's stated purpose.
Q: Can a "fiscal reservoir" and strategic investment coexist?
A: They are difficult to pursue together. A fiscal reservoir means storing revenue when it is plentiful and drawing on it in difficult times. Strategic investment, by contrast, means spending aggressively now on AI, semiconductors and the like. One is storage, the other withdrawal. The directions are exactly opposite. If both are needed, the accounts must be separated. Putting them in one pocket and claiming to pursue both stabilization and strategic investment is contradictory.
Q: What safeguards does the Future Response Fund need?
A: Rules have to come first. Upper and lower limits on how much to accumulate should be set, and the conditions for withdrawing money should be defined strictly. Not all excess revenue should be routed into the fund simply because it exists. A certain portion should be automatically applied to repaying national debt or public funds. Writing the figures into law is one option. Using the structural revenue measure I mentioned, the portion collected temporarily because of the boom must be identified. The future fund is not money for the current government to spend freely; it is a mechanism for leaving fiscal capacity to the next generation.
Q: The government says it has restructured 107 trillion won in spending.
A: Restructuring has to be judged by results. What if someone on a diet boasts about cutting out three bottles of cola while drinking ten more bottles of cider? They have consumed seven more bottles of soda. The current spending overhaul amounts to saying, "I cut the cola, so the diet worked." What matters is not how many existing programs were eliminated. You have to look at how the total of discretionary spending actually changed, including new programs. It is hard to credit large-scale restructuring when total spending has been raised 12.8% and only the programs that were cut are tallied up.
Q: Is the overhaul of education grants proper restructuring?
A: The structure that over-allocates education funding does need fixing. With the school-age population falling sharply, a system in which education grants rise automatically whenever internal tax revenue increases cannot be sustained. But moving money earmarked for education into another account, such as the Future Response Fund, and spending it there is not spending restructuring. Real reform would examine whether the total amount allocated to education is appropriate in the first place. Demand for health care and long-term care will keep rising as the population ages. If the demographic structure has changed, the structure of fiscal allocation has to change too.
Q: Should local government grants be revised the same way?
A: They should be viewed differently from education grants. Local governments face growing fiscal needs because of population aging and regional decline. Local grants are also tied to the fiscal autonomy of local governments. There is no guarantee that central government allocation is more efficient. That said, a structure in which enormous resources flow automatically to local governments whenever revenue rises abnormally is not appropriate. A separate set of principles for handling excess revenue should be established, and spending should take place within that framework.
Q: Should state-owned enterprise debt also be treated as a fiscal risk?
A: When the government assigns policy projects it should be carrying out itself to state-owned enterprises, the cost is not recorded as national debt in the near term. That does not make the burden disappear. Suppressing electricity rates piles up debt at KEPCO, and assigning policy housing projects to Korea Land & Housing Corporation increases LH's debt. Eventually it comes back as a burden on the public. Looking only at central government debt makes it difficult to gauge actual fiscal risk properly. Public institution debt, guarantees and other items that could convert into government obligations must be examined alongside it.
Q: The managed fiscal deficit is projected to exceed 100 trillion won in 2030.
A: That is what worries me most. Even the 100 trillion won figure assumes that revenue comes in quite strongly. It also assumes that high profits at semiconductor companies are sustained to some degree. But what if chip prices fall or global AI investment slows more than expected? Revenue could shrink far faster and the deficit could grow beyond 100 trillion won. We may be at the peak of a semiconductor super-cycle right now. Revenue recorded at the peak must not be misread as the normal capacity of our public finances.
Q: What is the fiscal line that must be held through 2030?
A: A national debt ratio of 53%. Since the fiscal rules use 60% as the reference line, the idea is to leave a buffer of about 7 percentage points. If debt is pushed to the limit in normal times, there are no cards left to play when a recession or financial crisis arrives. The bigger problem comes after 2030. Pension, medical and long-term care spending will begin rising in earnest as the population ages. We have to think about the next 20 years. This is not the time to spend all the fiscal capacity created by the chip boom. A financial crisis can be contained with fiscal resources. But if public finance itself falls into crisis, the last line of defense is gone.
He is
Born in Seoul in 1966, he graduated from Seongdong High School and the Department of Economics at Seoul National University before earning a doctorate at Yale University in the United States. After serving as a research assistant professor at Humboldt University of Berlin in Germany, he worked as a research fellow at the Korea Institute of Public Finance and as head of the revenue estimation team and tax analysis officer at the National Assembly Budget Office, earning a reputation for combining theory with practical experience. He has been a professor in the Department of Taxation at the University of Seoul since 2012 and has served as president of the Korean Association of Public Finance since April this year.
