Has the pass-through of the US dollar to US imports changed over time?

This study investigates whether the passthrough from the US dollar to the price of US non-oil imports has changed over time, secondly whether this question can best be studied using static or dynamic modeling. Two main conclusions can be made:

• The exchange rate pass-through changes over time, why a linear model allowing the pass-through to vary over time is preferable to log-linear models that assume constant pass-through
• A model using three lags of both dependent and independent variables have a higher explanatory power (adjusted R²=0.74) than both linear (adjusted R²=0.64) and log-linear models (adjusted R²=0.61) that only include a single lag of the dependent variable but no lags of the independent variables.

Diagram 1 and 2 below show how the exchange rate pass-through has changed over time for both linear (variable pass-through) and log-linear models (constant pass-through) that use only a single lag of the dependent variable. From the two diagrams the common finding in literature that the pass-through has decreased cannot be accepted unconditionally. For the period studied the exchange rate pass-through has varied, and while more data is needed to establish this, then there is some evidence of cyclicality in the pass-through.

Diagram 1: short-run pass-through                              

Diagram 2: long-run pass-through

 
You can read the rest of my analysis here: bit.ly/2jU1aO6

Forecast accuracy vs. forecast value when modeling the Dollar-Pound Sterling since the 19th century

Findings

This study estimate purchasing power parity (PPP) models of the US Dollar-Pound Sterling based on data from 1861-2014, after which the statistical accuracy and economic value of these forecasts are compared. This study finds that 1) the estimated model beats a random walk model in-sample and 2) that both the accuracy and value of out-of-sample forecasts fall as the forecast horizon increases.

Model developed

The developed models predict estimated future changes in the USD-GBP 1, 4, 8, 12 and 16 years ahead based on gaps between the spot and PPP-implied exchange rate. Table 1 and diagram 1 below show that all five models are statistically significant and have intuitive coefficients: an increase in the fundamental exchange rate (ft) over the spot exchange rate (st) leads to a future increase in the spot exchange rate across all forecasting horizons.

Table 1: model findings

Diagram 1: coefficients and forecast errors across different forecast horizons

To ensure congruency all five models were estimated using heteroskedasticity and autocorrelation-consistent standard errors. Except for the model forecasting four years ahead, then the parameter stability of other forecast horizons was deemed sufficiently stable to use for forecasting.

Forecast accuracy and value

The statistical accuracy and economic value of the estimated models were analyzed across the different forecasting horizons. The economic value of forecasts was examined using the directional accuracy and time value-adjusted root mean squared forecast error (RMSFE), and a stylized finding is that both the statistical accuracy and economic value of the estimated models decrease as the forecasting horizon increases. 

Read the full paper here: http://bit.ly/2qYoxdl

Capital mobility and the Feldstein-Horioka Puzzle

I recently completed an empirical research paper on the Feldstein-Horioka Puzzle (“FHP”) for my class Quantitative Global Economics and wanted to share my findings. One of my main findings can be seen in the diagram above. A higher savings coefficient indicates low capital mobility (0 is perfect capital mobility and 1 is full capital immobility).

The FHP is the empirical finding by Feldstein and Horioka (1980) that international capital mobility, when measured as the cross-country correlation between savings and investment rates, is rather low, despite economic theory arguing that capital is highly mobile.

My paper carries out two analyses: the original FHP is examined using the same methodology as the original paper but using different sample periods and country groups. Secondly, I examined whether the quality of economic institutions is correlated with capital mobility when measured as the difference between national savings and investment rates, something that previous research has not analyzed.

Findings

The first analysis of this paper used the original OECD country group that Feldstein and Horioka used and found that international capital mobility has increased from the 1970s until 2009, after which capital mobility has decreased. Using the same method but 193 countries it was found that capital mobility increased until 1999, after which fell.

The second part of this paper found that the only significant determinant of capital mobility during 1995-2014 was the openness of trade regimes, and that capital mobility is not significantly correlated with time after controlling for economic institutions. It should be noted, however, that the findings were not robust when analyzed separately for the first and last ten years of the sample period.

You can read the full paper here: https://goo.gl/HsVDcm 

Determinants of Chinese local government bond spreads

How come that Chinese local government bond yields are different from Chinese central government bond yields - despite that the central government guarantees local government debt?
My hypothesis is that provincial economic conditions should not matter for local government bond spreads given that the central government guarantees local debt. To test my hypothesis I analyzed whether provincial economic conditions can explain the central-local spreads of 2,680 Chinese local government bonds issued during 2015-2017 - bonds equal to 21% of Chinese Q3 2017 GDP.

My findings show that some indicators of provincial economic conditions explain local government bond spreads, although this finding is against my hypothesis. You can read the whole paper here: https://goo.gl/5SY5d1.

The below photo is from University of Pennsylvania's Symposium on Contemporary China, where I had the opportunity to present, and get feedback on, my research.

Chinese Local Government Debt: Reforms

This briefing is the second in a series of two that aims to introduce the non-specialist reader to Chinese local government debt. The first briefing shed light on why local governments in China borrow and why it is a problem, this article will discuss the reform wave that began in 2013 and the final briefing will explore the determinants of local government bond yields. This paper will firstly present its conclusion and after that discuss the extent to which recent policies solve the causes and problems of local government debt in China that was identified in the first briefing.

1. Conclusion

The reform wave that began in 2013 has introduced policies addressing both why and how local governments in China borrow. While a range of fiscal reforms have been implemented, then they do not address the root cause of excessive borrowing: local governments face unfunded fiscal mandates, something that is reflected in a rising expense-to-revenue ratio. Exacerbating the inconsequential fiscal reforms is that local governments will be unable to rely on land sales for much longer, something land reforms have left unaddressed. More positively, a large loan-for-bond swap program has helped increase transparency for local government finances, although the central government has reduced the pace of reform.

2. Recent reforms related to local government debt

In 2013 a new reform wave aimed at the fiscal and debt situation of local governments began. Broader fiscal reforms began in 2013 (source), and reforms directly related to local government debt began in 2014 (source), and both types of reforms have expanded continuously since. When evaluating the effectiveness of these reforms, it is useful to think about how the reforms address fiscal-local relations (“why” local governments borrow) and the structure of local government debt (“how” local governments borrow).

The previous briefing argued that the local government debt problem is caused by unfunded fiscal mandates and an unfunded post-2008 fiscal stimulus program, and is a problem due to the increase in the size of debt, the reliance on land sales and an increasingly opaque debt structure. While the post-2008 fiscal stimulus program pushed local government finances towards unsustainable levels, then it was a temporary policy and will not be discussed further.

3. Do recent policies solve the root cause of unfunded fiscal mandates?

Current local government fiscal issues can be traced directly to 1994 fiscal reforms, and a range of policies targeting central-local fiscal relations have been passed since 2013. (The specific policies are listed in appendix A). Unfortunately, so far it appears that recent policies account for little more than lipstick on a pig – the policies do not solve the unfunded fiscal mandates that local governments face – something that the State Council recognizes by stating that “there will be no major change to the distribution of fiscal revenue between central and local governments” (source).

Diagram 1 supports the above argument: there is evidence of a deterioration in the expenditure-to-revenue ratio of local governments since the middle of 2016. While the 2015-2016 decrease in the ratio that includes extra-budgetary items might be cause for optimism, then the simultaneous increase in the ratio that excludes extra-budgetary items could indicate that fiscal items have merely been shifted onto the budget.

Diagram 1: local government expenditure-to-revenue ratio (source)

It could be argued that reforms have only been implemented very recently, and that effects are therefore yet to be seen. This briefing, however, argues that economic indicators responded almost instantaneously following the 1994 fiscal reforms, why fiscal reforms should already be visible in aggregate numbers (- assuming that state capacity today is similar to in 1994)

4. Do policies address the problems associated with local government debt?

Major problems associated with local government debt in China include the increase in the size of debt, the reliance on land sales and the opacity of debt.

Size of debt

Although debt ceiling policies have been issued (source) then local governments will not be able to stick to such policies unless their expenditure-to-revenue ratio decreases. One factor that should push the expenditure-to-revenue ratio down marginally is that the interest rate on new bonds issued by local governments (discussed more later) generally is lower than for bank loans or shadow bank financing (source). Issuing bonds with lower interest rates, however, cannot be expected to reduce the expenditure-to-revenue ratio to a sustainable level by itself.

Reliance on land sales

Despite “land sales account[ing] for about 30 percent of local government revenue in 2016” (source) then recent land reforms do not address how local governments will overcome expected future decreases in land sales revenues. Land sales revenues are bound to decrease as a result of the average sales price of government land decreasing (source) and as the supply of land is predicted to be exhausted by 2021 (source). The reason for this is that an increasing share of the land being sold by local governments is being used for infrastructure, rather than e.g. industrial, commercial or residential purposes. Land used for infrastructure sells for much lower prices than other types of land (source).

Opacity of debt

More consequential, though, are the policies addressing the opacity of Chinese local government debt. As a result of bans on borrowing, local governments have sought financing outside of regular banking channels, with one example being local government financing vehicles (LGFVs). Two policy trends are addressing the opacity of local government debt: (i) off-balance sheet liabilities being transformed into on-balance sheet liabilities, and (ii) loans being swapped for bonds.

(i) A 2014 policy (source) recognized around two-thirds of LGFV debt – amounting to 17% of GDP – as being explicit local government debt (source). This one-off policy has allowed for a better understanding of the de facto size of local government liabilities, and can be expected to provide a more accurate classification of future liabilities if adhered to going forward.

(ii) Since 2014 local governments have been allowed to issue bonds (source), and are required to swap both LGFV liabilities and regular local government bank loans for publicly listed bonds. The bonds can be issued with fixed coupons and 1-, 3-, 5-, 7- or 10-year maturities (source). Government policies detailing how local government bond crises would be handled further increase transparency (source). While the crises-handling policies state that the central government would not bail out local governments, then this is not considered credible by mainstream China analysts. Analyst consensus appears to be that while the central governments might let a small number of LGFVs default, then it is not expected that the central government would allow widespread defaults. So far policy implementation has been rather successful: debt arising from loan-for-bond swaps has increased from 0% of GDP in 2014 to 4.5% in 2015 and 10.8% in 2016 (source). The IMF “assumes that future financing needs will be increasingly met by bond issuance, in line with the authorities’ plan to replace all local government debt with bonds” (source) by 2018 (source), something reflected in diagram 2 below.

Diagram 2: stock of outstanding local government bonds (source)

This briefing argues that the move toward on-balance sheet bonds is an effective way to increase the transparency of local government finances for a couple of reasons. Firstly, publicly listed bonds should allow the central government to leverage the collective insights of financial markets to monitor local governments. Additionally, whereas off-balance sheet liabilities embody uncertainty (that cannot be quantified) due to poor information regarding the size, quality, and interconnectedness of liabilities among other factors, then listing local government debt will allow for better quantification of risks.

One negative point about the bond swap program is that the central government is dragging its feet. The central government is trying to limit local government bond issuance due to worries about excessive borrowing, resulting in a 48% year-over-year drop in Q2 2017 local government bond issuance (source, based on the four-quarter moving average). The impact of the debt ceilings is reflected in the 2016-2017 local government bond issuance seen in diagram 3 below. Ironically, and as discussed above, the bond swap program cannot solve how much local governments borrow, only how they do so. This briefing argues that preventing local governments from raising financing in public markets could just force the local governments to return to shadow banking. Despite the feet-dragging then there are no indications that the bond-swap program will be discontinued.

Diagram 3: local government bond issuance over time (source)

5. Appendix A: reform policies

Policies that address the root cause of unfunded fiscal mandates:

• VAT and resource tax reforms (source, source, and source)
• Procurement and outsourcing of public services, including public-private partnerships (source, source and source)
• Requiring local governments to have balanced budgets and banning off-balance sheet finances (source)
• Improving central-local fiscal transfer mechanisms (source and source)
• Aligning generous public-sector pensions with private-sector standards (source)

Bond-swap program policies:

• First policy discussing local government bonds (source)
• Local bond market established (source)
• Specific-purpose bonds launched. These types of bonds will be repaid using only revenue from a specific project (e.g. a toll road) (source)
• Debt ceiling introduced (source)
• Further policies on local government bonds (source)
• Further policies on specific-purpose bonds (source)
• Guidelines on how to manage local government bond defaults (source)
• Debt ceiling regulations relaxed slightly (source)
• Crackdown on off-balance sheet borrowing (source)

Why is it difficult for China’s central government to make local leaders listen?

The ancient Chinese proverb “the emperor is far away and the mountains are high” signify the perennial power struggle between China’s central leadership and local officials and rings true even today. Due to having a history of regional government-led separationist movements, and a deep-seated national fear of chaos, China’s central government holds tighter onto power than in other large countries. The central government, however, often find – and accept – that local leaders slip out of its policy iron-fist. Although many assume that China exemplifies harsh authoritarianism, then leading scholars argue that China is actually a fragmented authoritarian state (link): local leaders are often not listening to the directives coming from the capital. Or according to a popular Chinese saying: “policy from above, opposition from below”.

Yet why is it so difficult for Beijing to control its localities? While many reasons undoubtedly exist, then China’s size, lack of effective checks and balances and information asymmetry are here argued to play large roles. China’s size – the average population of China’s 31 provinces is 44 million – is one reason why it is difficult for the central government to effectively monitor local politics and ensure compliance with central directives. But in contrast to other large countries such as the US, then China does not have effective checks and balances. The monopoly on power that the Chinese Communist Party has, ironically, makes it more difficult for it to control local leaders. Whereas democracy helps establish effective checks and balances, then the ultimate authority at the local level in China – the Communist Party – is supposed to monitor itself, creating conflicts of interest. Finally, as a result of the country’s vast size it is challenging for the center to understand local conditions as deeply as local officials, why the central government at times allow localities to deviate from central policy to better suit reality on the ground.

Chinese Local Government Debt: Introduction

This briefing is the first in a series of two that aims to introduce the non-specialist reader to Chinese local government debt. This briefing aims to shed light on why local governments in China borrow and why it is a problem and the second briefing will discuss the reform wave that began in 2013. Before proceeding, then local governments are here defined as the lower four of five levels in the Chinese political administrative structure: central, province, prefecture, county, township (source).

1. Why do we care about local governments in China in general?

One major reason that local Chinese governments are interesting is that most fiscal responsibility is held by local governments in China. More specifically, in 2015, 86% of government expenditure occurred at the sub-national level, while the sub-national government received only 54% of total government revenue (source). Two points arise when looking at these numbers. Firstly, China is much more decentralized in terms of both revenue and expenditure than all other major economies. In OECD countries, the average subnational share of revenue was 65% in 2015, whereas the share of expenditure was 31% (source). Secondly, the gap between subnational revenue and expenditure is larger in China than in other large economies. Whereas the 2015 local government fiscal deficit was 32% of GDP for China, then the average OECD country had a local government fiscal surplus of 34% of GDP. The below figure provides a more detailed overview of revenue and expenditure at different government levels in China.

Diagram 1: China is fiscally decentralized (2015 data) (source)

2. Why do Chinese local governments need to borrow?

The previous section gave a hint as to why local governments in China borrow: they have to spend more than they “earn”. Yet the revenue-expenditure gap is not the whole story. While there are likely many reasons for the increase in local government debt, the 1994 fiscal reform and post-2008 fiscal stimulus are two main causes. Examples of other frequently mentioned reasons include 1) that the central government implicitly guarantees local debt, leading to moral hazard among local governments, 2) that economic growth is emphasized in the evaluation of local politicians, leading to debt-fueled growth, and 3) that local officials can influence banks, allowing local governments to access financing on non-market terms. This briefing argues that these three factors, while important enabling contextual factors, are not root causes as they have existed for decades in China.

2.1. Long term: 1994 fiscal reform

The massive fiscal deficit faced by local governments in China can be traced directly to fiscal reforms implemented in 1994. While China experienced real economic growth that averaged around 10% per year between 1979 and 1993, government revenue as a share of GDP fell from 28% to 12% during the period (source). The fall in government revenues led the government to implement major fiscal reforms in 1994, something that enabled the central government in Beijing to increase its share of total government revenue from 22% to 56% during 1993-1994 and nearly double the ratio of government revenue to GDP in the following 20 years (source). The 1994 reforms achieved its goal of fiscal centralization, although local governments until today face “expenditure responsibilities [that] exceed their assigned revenues” (source), why the local governments have had to find ways to fill their revenue gaps. Local governments have plugged their funding gap by borrowing from banks and shadow banks, and selling land (often by expropriating land from citizens and compensating them at below-market rates).

Diagram 2: impact of the 1994 fiscal reform (source)

2.2. Medium term: post-2008 stimulus

During 2008-2010 the Chinese government decided to implement major fiscal stimulus programs worth almost $600 billion (source) to counteract the financial crisis, with around three-fourths of the money going to local governments. While this initially might sound positive for local governments, despite “the significant outlays, the money the center [gave was] nowhere near enough to fund all the local projects” that it approved (source). Despite the stimulus package resulting in local governments taking on more debt through a range of direct and indirect channels, then most of the money was spent on infrastructure and other public goods – projects on which “the income-earning potential of the projects is limited” (source). According to one estimate, local government debt increased by around 60% in 2009 (source).

3. Why is local government debt a problem in China?

Although the aggregate national debt burden is not a cause for huge concern at the current debt levels (source), then this briefing argues that the rapid increase, geographical variation, reliance on land sales and opacity of debt are concerning.

Increase in size

The increase in size of local debt, and the ability of local governments to cover such debt is a rather straight-forward story: the debt burden has increased significantly as both a share of GDP and as a share of local government revenue, however measured.

Diagram 3 and 4: increase in size of debt (source)

The national average of local debt, however, covers a broad range of provincial debt levels, as seen in diagram 5 below. The broad variation in debt burden could easily resemble the variation in debt across a sample of countries, making this author ponder whether a potential local debt crisis in China might be idiosyncratic or systematic across only a subset of provinces, particularly given clustering in the economic profiles of Chinese provinces.

Diagram 5: provincial variation in debt burden (source)

Reliance on land sales

Further supporting the above narrative is that local government revenue is heavily reliant on land sales; according to one estimate, “land sales accounted for about 30 percent of local government revenue in 2016” (source). The supply of land available for the local government to sell, however, is expected to be fully exhausted by around 2021 (see diagram 7 below).

Diagram 6 and 7: reliance on land sales for local governments (source 1; source 2)

Increased opacity

Local governments in China are “legally prohibited from borrowing or running deficits” (source) following the 1994 fiscal reforms. So how do local governments escape this oxymoron? While local governments have obtained “much of their debt in the form of bank loans”, then this is not a cause for concern as direct local government debt only increased from 16.5% of GDP in 2006 to 18.6% of GDP in 2013 (ADB, 2014). The main concern comes from opaque financing accessed through shadow banking. Following the stimulus plan announced in 2008 the central government authorized local governments to access debt through formal channels (source). As formal financing channels were not sufficient, local governments increasingly sought credit outside the formal banking system.

One common shadow financing structure is a local government financing vehicle (LGFV), which are enterprises set up by the local government and implicitly guaranteed by the central government. As LGFVs are not officially part of the government they can borrow on behalf of the local government. While the central government is no doubt aware of LGFVs, LGFVs have been overlooked as they have allowed local governments to fulfill their fiscal commitments. The next briefing will include a discussion on the decreasing reliance on LGFVs of local governments.

The increased reliance of local government finances on shadow banking is an issue for multiple reasons, some of which are described below. While estimates of the size of shadow banking are available, it is difficult to have a high degree of confidence in any single estimate given that estimates of the size of shadow banking has a large variance (source). An examination of estimates of shadow bank financing accessed by both government and corporates in 2013 found figures ranging from 8% (Standard Chartered) to 82% (JP Morgan) of GDP, with the average figure being around 50% (source). Another issue is that shadow banking can be argued to embody uncertainty and not risk: although case studies on shadow financing structures can be developed, then it is difficult to develop a credible view on the macro-level risk features of shadow banking. Finally, while LGFVs are widely assumed to be implicitly guaranteed by the central government, LGFV bond yields reflect provincial characteristics such as reliance on real estate, corruption, liquidity indicators and macroeconomic variables (source), indicating that markets do not view the central government implicit guarantee as fully credible. The central government has indeed recently argued that it is willing to let LGFVs fail; while analysts believe that the central government might allow a small number of LGFVs to fail (source), consensus is that widespread defaults would not be allowed.

4. Conclusion

The problem of local Chinese government debt can be traced back to the 1994 fiscal reform, resulting in unbalanced fiscal mandates being forced upon local governments. The problem accelerated following the 2008 fiscal stimulus program, which led local governments to tap the shadow banking system to fulfill a centrally underfunded fiscal stimulus program. While the average national level of local government debt is not worrying by itself, the growth rate, regional variation, reliance on land sales and opacity could be sources of issue in the future.