Hong Kong has an enviable problem. Over the years, the Hong Kong government has accumulated a huge amount of reserves (roughly HKD 4000 billion), way beyond the amount needed to defend the Linked Exchange Rate System and keep the government solvent (by the way, the government is one of the rare ones that have no external debt). Other than relatively stable sources of revenue like income taxes, the government benefits mostly from land sales and the transaction-driven stamp duties. When such an amount of wealth is sitting around earning a modest, if not disappointing, rate of return, any economist worth his or her salt would ask the question: is this the most efficient allocation of resources?
Recently, Joseph Yam, the former head of the Hong Kong Monetary Authority and now a member of the Executive Council under the new government of Carrie Lam, provided an answer to this question. His answer can be interpreted as a qualified “no,” and yet it was enough to cause a stir in Hong Kong. The responses were mostly negative, with people criticizing Yam for his obvious change of stance and suggestion to “throw away” the wealth of the Hong Kong people.
I am a bit less critical and I think that Mr. Yam is pointing the discussion towards the right direction, to say the least.
The Elusive and Much-Abused Multiplier
Fiscal policy in Hong Kong is supposed to follow a “rule,” which is Article 107 of the Basic Law which states that “(the) Hong Kong Special Administrative Region shall follow the principle of keeping expenditure within the limits of revenue in drawing up its budget, and strive to achieve a fiscal balance, avoid deficits and keep the budget commensurate with the growth rate of its gross domestic product.” As an economist, I find the Article unnecessarily restrictive. No matter if we are talking about the whole economy or just an individual, it makes little sense to always avoid spending more than the amount taken in. When facing a war or some large-scale natural disaster, a government deficit is usually the right response. When arranging a wedding or holiday in Europe, dipping into savings is more feasible than refusing to spend more than your current earnings.
The last clause of the Article is especially vague. The usual interpretation is that the size of the budget should have a similar rate of growth as the gross domestic product (GDP). Mr. Yam disagrees, and he reads into the word “commensurate” and suggests that the size of the budget should instead go in the opposite direction of the GDP. When the economy is booming, the budget should be reduced to avoid overheating. When the economy is sluggish or even shrinking, the budget should be increased to stimulate growth. What should the government spend on as a form of stimulation? Mr. Yam recommends that “priority can be given to those items with higher multiplier and productivity-enhancing effects.”
Where are these higher multiplier effects? Multiplier effects are infamous in economics as being theoretically suspicious and empirically elusive. Usually wrongly attribute to John Maynard Keynes (it was his friend Richard Kahn who first came up with the idea), the concept describes how a certain change in government budget (purchases, taxes, or transfers) affects the whole economy. If an increase of $100 in government purchases leads to an increase in GDP by $150, the multiplier effect is then 1.5. If GDP only increases by $50, the multiplier effect is weaker at 0.5. How large is the multiplier effect in practice? There are numerous empirical studies on this, and the answers vary. Most studies point to a positive and small number.
Building new things is not the only answer. Indeed, there are tax cuts and transfers that may have “high multiplier and productivity-enhancing effects.”
Multiplier effects are tricky as they vary across projects and over time. The famous example of digging a hole and then filling it is of a poor use of resources, and large infrastructure projects are notorious of being wasteful and inefficient. Investment in education or technology, while widely believed to be productivity-enhancing, may suffer from misallocation (without the guidance of market prices) and rent-seeking. Other than conducting case-by-case studies, there is no reliable way to categorize some projects as “good” and others as “bad.”
Do not be so easy to believe an expert who tells you “the government should always invest more in education” or that “education should take up X percent of government spending.” Invest in what specifically? How do we know that education spending always and everywhere brings a high return? Also, be skeptical next time the government releases a consulting report claiming that an infrastructure project is going to bring benefits that surpass costs. How were those benefits estimated? How many indirect effects were sneaked in without good reason? One needs economic reasoning and empirical evidence to verify such claims.
The Relevant and Much-Ignored Crowding-Out
Multiplier effects also change with the state of the economy. Imagine that there are only 10 people in the economy and that they are all fully occupied with their jobs. Now, the government plans to build a bridge, and the project needs a certain number of working hours to be completed on time. To attract people to change jobs, wage will increase until enough people work for the government. With the economy already at full capacity, the infrastructure project mostly leads to higher wages and fewer other goods being produced in the private sector.
Hong Kong now has a low unemployment rate of 3.3 percent that is envied by most other countries, and it is not far-fetched to argue that we are near or at full employment. It is exactly during this period of low unemployment that the Hong Kong government started several large construction projects (Express Rail Link, Hong Kong-Zhuhai-Macao Bridge, and the recent Kai Tak Sports Park, just to name a few). The large increase in demand for construction labor, raw materials, equipment, and other inputs, when combined with supplies that are not perfectly elastic, pulls up prices. One obvious consequence is the higher construction cost of private residential housing which contributes to housing prices that are already at record highs.1
“Job creation” is always a selling point for infrastructure, and the right question to ask is whether “job creation” is merely “job change” in disguise.
Government Spending Is Not the Only Answer
A common misconception of fiscal stimulation is that it is equivalent to government spending in infrastructure and other public investment. While it may be true for the New Deal, it is far from the truth for more recent cases. A recent study shows that the spending from 2007 to 2009 for OECD countries as a response to the global financial crisis consisted mostly of tax cuts and transfers.2 The researchers also found that, in theory, these transfers’ multiplier was much larger than the purchases’ multiplier, once the use of the resources of the latter is taken in account.
It is unclear whether Mr. Yam mostly had infrastructure projects or other public investment programs in mind when he talked about government spending, but I think he probably agrees that building new things is not the only answer. Indeed, there are tax cuts and transfers that may have the “high multiplier and productivity-enhancing effects” that he desires. Let me mention three that I have discussed in more detail on other occasions:
1) I have long advocated for returning a portion of the reserves to the general public in the form of cash. It is administratively simple and will avoid most of the waste involved in government spending. Motivated by self-interest, money in the hands of the people is more likely to be spent in beneficial and effective ways. No one knows better than ourselves what we are most willing to pay for.
2) For a more long-term policy, I have also proposed establishing a system of longevity insurance for people over some “old age” cutoff, and at the same time eliminating all the other clumsy and complicated welfare programs. It is hard to plan ahead when it is uncertain when you will die, and longevity can be seen as a “misfortune” that eats into retirement savings. The government can make use of its ample reserves to compensate people who live longer, at the expense of people who die sooner.
3) Reducing or even eliminating stamp duties and other incomes related to the stock or property market will reduce the volatility and unpredictability of government revenue. Such fickle revenue is the major reason why the former Financial Secretary John Tsang repeatedly got the budget estimate wrong in the past. The unpredictability has become an excuse for the government not to commit to any long-term programs, and instead to just accumulate such “windfalls” until it ends up with a large stockpile of reserves.
The Hong Kong government should at least consider the possibility — especially when Hong Kong has such a low unemployment rate — that such tax cuts and transfers may be better ways than spending to allocate its reserves.
The discussion of fiscal policy in Hong Kong has long suffered from the problem of being too vague and too broad. We have spent too much time debating the forest (“We should spend more on infrastructure and education”, “We should keep government spending at X percent of GDP” or “Information technology badly needs more investment from the government”), and we have not paid enough attention to the messy details of the trees. Mr. Yam has corrected that imbalance in the discussion by a little, but it is still not enough. Individual projects should be evaluated by scientific, data-driven, and evidence-based procedures, and the Hong Kong government should be more transparent with the evaluation processes and the underlying assumptions behind their claims. With no more baseless proclamations and obscure consulting reports, it will be more likely that the plentiful reserves of Hong Kong will move toward a more reasonable level and achieve more efficient allocation.
1. Some crowding effects are subtler. If the Express Rail Link is indeed successful, and people from both the mainland and Hong Kong opt for the train, then it will cast doubt on the optimistic estimation of the demand for the Third Runway.
2. Oh, H. and Reis, R. (2012). Targeted transfers and the fiscal response to the Great Recession. Journal of Monetary Economics, 59, S50-S64.