2005 all over again? Given another exit from a US dollar peg, comparisons with the de-pegging in 2005 are natural. Then, the PBoC started the ball rolling with a small one-off revaluation as part of a 20% appreciation over a three-year span. The current appreciation is likely to be as measured and gradual as the last one, in our opinion.
Not quite - the growth outlook this time is different: The backdrop this time around is very different. In 2005, growth was robust throughout the world, while the US dollar was weak and about to get weaker still. This time, there are only a handful of countries around the world where growth has been robust enough to withstand the withdrawal of policy stimulus. The US dollar reversed course at the end of 2009 and has performed very well since then. The renminbi is therefore now appreciating relative to a strengthening currency rather than a weakening currency, as was the case in 2005. In addition, the euro's weakness due to sovereign and banking risks means that the renminbi is now stronger against its biggest export markets. This, in itself, makes the PBoC's policy move a significant event. Furthermore, the economic significance will surely increase if this regime is the harbinger of a more flexible exchange rate regime in future.
Seven Reasons to De-Peg: Peeking Back at 2005
Many of the arguments made in 2005 for a move from a fixed exchange rate to a flexible exchange rate for the renminbi still resonate today. Perhaps the best example of this is a list of seven arguments for the renminbi to move to a more flexible regime, written by Jeffrey Frankel (NBER Working Paper No. 11274, April 2005). These are as follows:
• Appreciation would help control inflation amid fears of overheating.
• Foreign exchange reserves are more than adequate, and US Treasuries don't pay a high return.
• It becomes increasingly difficult to sterilise inflows over time, exacerbating inflation.
• Although external balance could be achieved by expenditure reduction, i.e., by raising interest rates, two policy goals of internal and external balance need two policy tools (e.g., real exchange rates and interest rates).
• Real exchange rate flexibility is much easier to achieve through nominal exchange rates than through prices for a large economy like China.
• The experience from other emerging markets supports exiting a peg when times are good and a currency is strong.
• From a longer-run perspective, prices of goods and services in China are low.
Even a simple glance at these reasons suggests that the underlying reasons for China to move to a more flexible exchange rate regime are at least as compelling now as they were in 2005, and probably more so now that its economic outperformance is so stark.
As we point out, there are important differences between the backdrop in 2005 and now. With the global economy yet to convince everyone of the sustainability of the economic recovery, the global growth picture is very different from the rosy one in 2004. The other big difference is the US dollar. Even the modest renminbi appreciation is likely to be more significant this time around since it is relative to a US dollar that is strengthening (as opposed to an appreciation against a weak and falling dollar in 2005).
Taking no chances: Ever since Chinese policymakers started their resolute campaign against property speculation, markets have been worried about a policy error that could push such anti-speculation measures too far. Mindful of fears of a growth slowdown in China, monetary policy is likely to compensate for renminbi appreciation by delaying tightening or even by further easing on some measures. Our China economist, Qing Wang, previously expected just one policy rate hike in 2H10 and an adherence to the Rmb7.5 trillion credit ceiling for 2010. He now expects no rate hikes in 2010 and raises the possibility of an increase in the credit ceiling (see again China Economics: Renminbi Exits from USD Peg and Returns to Pre-Crisis Arrangement). Monetary policy is thus playing a much more accommodative role as interest rates and credit constraints together account for the bulk of monetary transmission into the real economy.
But monetary policy accommodation shows that the ‘trilemma' is alive and well: The trilemma is a three-way dilemma whereby policymakers can choose only two out of a trio of pegged exchange rate, free flows of capital and independence in monetary policy. The traditional approach of policymakers who want to have their cake and eat it has been to introduce capital and credit controls. Such a system of controls in China (including intervention in FX markets, sterilisation in domestic money markets and credit constraints) has kept capital inflows from influencing domestic credit but has led to an accumulation of a massive quantity of foreign exchange reserves. Managing these controls and reserves, however, gets more difficult as time passes and they do not represent a sustainable solution. The textbook model can lead one to believe that the trilemma exists only for exchange rate regimes that are fully fixed. However, this is not the case, in our view. Even somewhat flexible exchange rate regimes can produce the same constraints, leading to a policy trade-off. In the case of the Reserve Bank of India, for example, the reliance on capital inflows has meant that the RBI has had to walk a tightrope between raising rates quickly, and excessive currency appreciation thanks to ‘perverse' inflows of capital.
The predictability of the new regime could make the trilemma worse: In theory, the switch from a fully fixed exchange rate to a slightly more flexible one should reduce the constraints of the trilemma. However, most market participants expect a modest appreciation in the renminbi against the US dollar, and this predictability could work to make the trilemma worse in the short run. How? The combination of a predictably appreciating currency and a monetary policy regime that has postponed tightening in order to support growth provides the investor with: (i) increasing returns for capital inflows purely from currency appreciation, and (ii) policy support for the economy and for risky assets. Raising rates in such an environment would invite ‘perverse' capital flows of the kind that the RBI has been wary of. Thus, rather than affording some respite from the trilemma, the predictability of the new regime may make the policy constraints of the trilemma worse.
Importing AAA liquidity: Capital flows into China may be exacerbated thanks to an unlikely source - the euro area. Sovereign risks from the euro area have probably prompted not just the ECB but also the Fed to stay on hold for longer, keeping the AAA liquidity regime intact. The fast-growing economies of the world, particularly the ones with fixed or relatively stable exchange rates, import this excess liquidity, and China is high up on that list.
In the near future, more capital inflows, higher reserves and lower yields are the likely outcome. As capital inflows increase, China's system of capital and credit controls will probably lead to an increase in the size of its holdings of foreign exchange reserves. We believe that at least some of these reserves are likely to find their way back into bond markets, keeping yields anchored. Despite an extremely volatile period in developed markets, US markets - and even euro area markets - remain a favoured destination thanks to their size and depth.
Two-way risk? Our AXJ currency strategist, Stewart Newnham, believes that the PBoC could alter the daily fixings of USD/CNY in a way that could create two-way risk in the market in order to curb the kind of near-arbitrage we have discussed above. He still expects the overall trend in the exchange rate to be lower but argues that the path could be bumpier if the PBoC indeed plans to generate such two-way risk. To truly test the thesis that de-pegging will lead to greater capital inflows, the volatility created by any such path of higher and lower fixings would have to be large enough to discourage investors from positioning themselves for an overall modest appreciation of the renminbi.
Further down the line, a more flexible and less predictable exchange rate regime could reduce the constraints of the trilemma and deliver a more balanced domestic and global economy. If a less predictable regime were to unfold, we believe that monetary policy would gain much more independence and the accumulation of reserves would likely slow down significantly or even reverse itself. Under these circumstances, risky or riskless assets in the developed world that have been the beneficiaries of recycled reserves would then have to offer better returns and higher yields to attract buyers. Exactly when forward-looking markets will begin to focus on this scenario is difficult to predict.
The DPJ government has published its new Fiscal Management Strategy (FMS, referred to in the debate so far as the ‘fiscal framework'). The FMS has a main text and a numerical appendix. The main text is long and vague; it restates known positions but does not clarify how much regulatory reform is needed to offset fiscal consolidation, how the tax system should be reformed, the impact of tax and spending changes on incentives, etc. However, the numerical appendix is particularly interesting. The latter is where the most concerns are found, after examining the numbers and the economic model behind them.
There are four key problems in the FMS: (a) a very large unspecified source of demand to close the output gap in the next five years; (b) lack of numerical backing for the fiscal control scenario outlined by the government; (c) counter-historical assumptions on the gap between interest rates and growth, leading to an insufficient target for the primary balance; and (d) an inherent tax-and-spend program that would require consumption tax hikes far in excess of the currently debated hike from 5% to 10%. Let us examine these points in detail.
1. Unspecified Demand Sources
First, even with spending control and a large amount of unspecified demand creation, there is still a fiscal gap to be filled. For example, in the ‘cautious scenario' that assumes accelerated increases in social security needs, the calculations presume (a) the spending freeze suggested in the main document, (b) the closing of the output gap by 2015. Even with a bit of increase of the current account surplus and some unspecified demand components, the calculations still show the primary balance for 2020 at a deficit of about 4% of GDP. So the cautious scenario is not enough to reach the fiscal goals. With the ‘growth scenario' that assumes no acceleration of social security needs, the calculations presume the same spending freeze, but also show a closing of the output gap two years earlier. Still, the primary balance is in deficit of 2.1% of GDP in 2020.
2. Lack of Quantification of the Fiscal Control Scenario
An even more revealing point comes in the footnote to Figures 1 and 2 of the numerical appendix. The figures purport to show what will happen if an extra ‘fiscal control scenario' is implemented. This scenario shows the economy reaching the targets mentioned in the main text, i.e., a primary balance of zero and a peak of debt/GDP ratio in 2020. However, the footnote says, "The fiscal consolidation scenario is not a calculation from the [econometric] model, but rather shows the goals of the fiscal consolidation in the fiscal control scenario."
This point is crucial. The lack of model-based calculations to generate results for the fiscal control scenario implies that there is some source of demand that is unspecified - on top of the unspecified demand that closes the output gap while fiscal spending is capped. How does the DPJ plan to achieve this scenario? What are the numbers and the model behind them?
3. Optimistic Target for Primary Balance Adjustment
Another problem emerges in the above-mentioned footnote. The very last sentence says, "The graph also assumes that the growth rate and the interest rate are equal."
This assumption that the interest rate and the growth rate are equal is counter-historical. In fact, the interest rate (average JGB yield on outstanding JGBs) has been about 1.4% above the growth rate. This point is crucial, because the gap between these two numbers determines how large the primary surplus must be, in order to stabilize the debt/GDP ratio. Had the calculations used the historical average of 1.4pp, then stabilization of the debt/GDP ratio would require a primary surplus of about 3% of GDP. With this more historical assumption on the difference of interest rates and the growth rate, the fiscal gap between the ‘cautious scenario' and the goal of stabilizing debt/GDP would not be 4% of GDP but 7%.
4. The Tax-and-Spend Strategy
The core of the DPJ argument is actually contained in a label in Figure 3. The label addresses the gap between the ‘cautious scenario' and the fiscal control scenario. The explanation says, "In the case that social security services are increased and the cost of this is shared by the populace, demand will rise, and it can be expected that the primary balance gap improve due to the increase of tax revenues." Unfortunately, neither the main text of the fiscal framework document nor the numerical appendix explains further. There is one potential explanation: the combination of direct spending and reduced anxiety about the future more than offsets the impact of lower disposable income, with the result of higher GDP and a lower deficit.
There are three key questions about this potential explanation: (a) Is it theoretically possible? (b) What do reasonable parameters in a Keynesian model suggest about the size of the effect? (c) How much of a hike of spending and taxes would be needed to close the deficit?
My answers to these questions are: (a) Yes, it is theoretically possible, although the actual answer depends on model parameters. (b) Each 1pp hike of spending and taxes would generate a maximum of 1/4pp reduction of the fiscal deficit, even on optimistic assumptions about the Keynesian multiplier. (c) The required tax-and-spend hike is 16pp of GDP, or the equivalent of a 32pp hike of the consumption tax (i.e., from 5% now to 37%). The calculations behind my answers are presented in the appendix in the full report.
5. The FMP Calculations Do Not Stick to the JPY44 Trillion JGB Issuance Ceiling
Another interesting, if less crucial, point is that the fiscal framework calculations do not include the target of holding new JGB issuance in 2011 to the same level as 2010. Both the ‘growth scenario' and the ‘cautious scenario' show the difference of spending and revenue in 2011 to widen to JPY49.2 trillion, or JPY4.9 trillion above that of 2010.
The apparent difference from stated policy (i.e., the policy of "not exceeding the current year's bond issuance", now slated at JPY44.3 trillion) is addressed by a clause on the front page of the numerical appendix: "These calculations...were created by the Cabinet Office [Naikaku-fu] and have not been approved by the Cabinet."
For full details, see Japan Economics: The Fiscal Management Strategy: Deep Dive, June 24, 2010.
Read Full Article »