International Monetary Fund (IMF)

Fiscally Driven Rebalancing Turns Out to Be Hard

The idea behind “fiscally-driven external rebalancing” is straightforward. If countries with external (e.g. trade) surpluses run expansionary fiscal policies, they will raise their own level of demand and increase their imports. More expansionary fiscal policies would generally lead to tighter monetary policies, which also would raise the value of their currencies. And if countries with external (trade) deficits run tighter fiscal policy, they will restrain their own demand growth and thus limit imports. Firms in the countries with tighter fiscal policies and less demand will start to look to export to countries with looser fiscal policies and more demand. This logic fits well with IMF orthodoxy: the IMF generally finds that fiscal policy has a significant impact on the external balance, unlike trade policy.* But it often encounters opposition, as it implies that the fiscal policy that is right for one country can be wrong for another. Many Germans, for example, think they need to run fiscal surpluses to set a good example for their neighbors. Yet the logic of using fiscal policy to drive external trade adjustment runs in the opposite direction. To bring its trade surplus down, Germany would need to run a looser fiscal policy. The positive impact of such policies on demand in Germany (and other the surplus countries) would spillover to the global economy and allow countries with external deficits to tighten their fiscal policies without creating a broader shortfall of demand that slows growth. So one implication of using fiscal policy to drive trade rebalancing is that there is no single fiscal policy target (or fiscal policy direction) that works for all countries. Budget balance for example isn’t always the right goal of national fiscal policy. Some countries need to run fiscal deficits to help bring their external surpluses down. That idea certainly encounters resistance. And as practical matter, the IMF’s latest estimates show that Europe hasn’t used fiscal policy to help facilitate its own internal adjustment.** The big external surpluses in the eurozone are in the Netherlands and Germany. Germany and the Netherlands also have a lot of fiscal space thanks to relatively low levels of public debt, and could safely run expansionary fiscal policies without calling their own fiscal solvency into question. And the countries with lots of external debt and limited fiscal space tend to be further to the south: Italy and Spain for example (I am over generalizing a bit here—Spain has more external debt than Italy and a bigger fiscal deficit, while Italy has a bigger public debt stock and less growth).*** The eurozone also runs a significant overall external surplus, has internal economic slack and has low interest rates—it could help bring global trade into better balance (and raise the global return on saving) with a more expansionary overall fiscal policy while also bringing its own economy closer to full employment. Win-win, at least in theory. In practice, though, the eurozone didn’t run an expansionary overall fiscal policy last year. The change in the structural fiscal balance was positive, but only just.*** And the fiscal contraction in Europe last year came from the external surplus countries: the Netherlands notably. Not from the former external deficit countries. Fiscal expansions in Italy and Spain provided the offset that prevented the consolidation in the Netherlands from giving rise to an overall consolidation in the eurozone. The pattern of fiscal consolidation in the eurozone is a bit different than what the IMF recommended. The IMF wanted fiscal expansion in the Netherlands, and fiscal consolidation in Spain, Italy, and others (generally at a pace of about 0.5 percent of GDP a year)—more or less the opposite of what happened. And what of Germany? Well, the IMF thought a year ago that Germany was doing a fiscal expansion—one that would bring its fiscal surplus down from about one percent of GDP (see paragraph 9 of the staff report, and the table on German general government operations on p. 8). But it turns out there wasn’t much of a structural fiscal expansion in Germany last year. The structural fiscal balance stayed in a substantial surplus. That’s the problem. It turns out surplus countries seem to like surpluses. They often aren’t willing to take policy action to expand demand. And the since the output of the eurozone's traditional deficit countries is generally constrained by weak demand, they tend to want to run more expansionary policies to boost their economies. The same basic point applies globally. For fiscal policy to drive global demand rebalancing, the external surplus countries around the world need to run more expansionary policies. That means the eurozone, Korea, and Japan, among others, should adapt more expansionary policies (along with Sweden, Switzerland, and Singapore). And of course the U.S. would need to adopt a more contractionary fiscal policy, one aimed at bringing the U.S. current account deficit down. But there isn’t much sign in the IMF’s data for 2016 that the surplus countries are willing to do meaningful fiscal expansions (and as I noted last year, in many cases the IMF hasn’t been willing to advocate fiscal expansions in its fiscal advice to major surplus countries). Korea’s structural surplus remains big (the IMF looks at a measure of the fiscal balance that includes the surplus in Korea’s social security system, which offsets the headline deficit) and didn’t change much in 2016. The latest WEO data suggests a (very) modest structural fiscal tightening in 2016, with more tightening in 2017. Japan has an ongoing structural fiscal deficit—its current account surplus comes from high corporate savings, not a tight fiscal policy. It has slowed the pace of consolidation from 2014 (thankfully) but its structural balance does't suggest that is doing much to support demand. It continues to rely heavily on net exports to support its growth. And, as for the United States…well, the President wants a big deficit-funded tax cut. **** *The IMF recognized in its 2016 external sector report that fiscal policy should play a role supporting balance of payments adjustment in many (but not all) “surplus” economies. See paragraph 28: “Countries facing stronger-than-warranted external positions and negative output gaps should primarily rely on fiscal policy to help close both domestic and external gaps, although the stimulus should be geared to support structural reform objectives. However, reliance on fiscal support depends very much on the availability of buffers. Korea, Sweden, Thailand, and the Netherlands appear to have room to ease in relative terms. Other countries, like Japan, where current fiscal space is more limited, will need to coordinate carefully the use of monetary and fiscal space with structural and income policies. Meanwhile, in the few cases where positive external gaps are paired with near zero or positive output gaps (Malaysia, Germany), monetary policy, if available as an independent instrument, should play a larger role (Malaysia). Those economies without independent monetary policy but with fiscal policy space (Germany), should use that space to finance growth-friendly policies that would support external rebalancing (including through an internal appreciation) with only a temporary and limited effect on the output gap." ** For the charts, Cole Frank and I used changes in the cyclically adjusted structural fiscal balance, as reported in the the IMF's latest WEO data base. There is an argument that the structural fiscal balance isn't the right measure, as interest expenditure has fallen (though you can debate this—interest paid domestically should support some spending), and the right measures for the fiscal impulse is the change in the cyclically adjusted primary fiscal balance. The IMF doesn't report that though (it reports the primary balance, but not the cyclically adjusted primary balance). Eyeballing the numbers, though, doesn't suggest this matters much, with the potential exception of Japan (where the debt really is held domestically). For Spain in 2012, I did use the change in the cyclically adjusted IMF primary balance as reported in the 2014 IMF article IV—as the structural balance includes the bank recapitalization bill (the "official" number implied a consolidation of around 4 percent of GDP). More generally, it should be noted these numbers are estimates, not the word of god. The 2016 numbers likely will be revised. *** Some additional throat clearing. The eurozone used to have an internal version of the world’s balance of payments imbalances. Some countries ran big surpluses, others ran big deficits. By and large, those deficits have disappeared, while the surpluses stayed big—hence the rise in the eurozone's surplus with the world. But the large deficits left behind a large stock of external debt (notably in Greece, Portugal, and Spain), and the fall in the deficit left many parts of the eurozone short demand. So an imbalances problem inside the eurozone turned into a demand problem inside Europe, and, without the deficits in the “south” the eurozone started running large external surpluses and exporting its savings to the world. **** U.S. Treasury Secretary Mnuchin has embraced the argument that fiscal reflation in the surplus countries can help address global balance of payments imbalances. I agree. But it is kind of hard to square that argument with the Trump Administration’s deficit raising tax proposals.

A long time ago I confessed that I like to read the IMF’s World Economic Outlook (WEO) from back to front. OK, I sometimes skip a few chapters. But I take particular interest in the IMF’s data tables (the World Economic Outlook electronic data set is also very well done, though sadly a bit lacking in balance of payments data).* And the data tables show the combined current account surplus of Europe and the manufacturing heavy parts of Asia—a surplus that reflects Asia's excess savings and Europe's relatively weak investment—remained quite big in 2016. China's surplus dropped a bit in 2016, but that didn't really bring down the total surplus of the major Asian manufacturing exporters. Much of the fall in China’s surplus was offset by a rise in Japan’s surplus. The WEO data tables suggest that net exports accounted for about half of Japan's 1 percent 2016 growth—Japan isn't yet growing primarily on the basis of an expansion of internal demand. And the combined surplus of Korea, Taiwan, Singapore and Hong Kong remains far larger than it was before the global financial crisis in 2008. The Asian NIEs (South Korea, Taiwan, Hong Kong, and Singapore) collectively now run a bigger surplus than China. As a result, in dollar terms—and also relative to the GDP of Asia’s trading partners—"manufacturing" Asia's combined surplus hasn’t come down that much over the last ten years. The size of the combined surplus of Europe and “manufacturing” Asia necessarily means that other large parts of the global economy need to run large deficits in manufactured goods. To be sure, barring an energy revolution, the big oil and gas exporters will necessarily trade oil for manufactured goods (and holidays), and parts of Asia and Europe equally will need to trade manufactures for energy. But the big Asian and European manufacturing exporters could not maintain surpluses of their current scale in the absence of a U.S. trade deficit in manufacturing that is as large as it was back in 2005 or 2006. There are only so many ways the global balance of payments can add up. While the surplus of key parts of the global economy haven't moved much, the nature of the financial outflows that channel the current account surplus of Europe and Asia (their savings surplus so speak) to the rest of the world has certainly changed. Setting a few countries (Switzerland, and perhaps Singapore***) aside, governments aren’t directly channeling funds abroad through the build-up of their reserves and the assets of their sovereign funds. Here is a plot of the growth in Asia’s official assets. My measure of official asset growth is mostly reported reserve growth, but is has been adjusted to include changes in countries' disclosed forward position and China's "other foreign assets." I also added in the rise in the non-reserve portfolio holdings held by the public sector (and I assumed China's portfolio outflows are from government institutions). This is a way of capturing sovereign wealth fund and public pension fund outflows. I did not add in the external lending of China's state banks. To be honest these adjustments are primarily for my own satisfaction (they all come from the balance of payments data—I am trying to avoid valuation adjustment these days). Looking only at reserves (adjusted for forwards and China's other foreign assets) would not materially change the picture. And here is the similar plot for Europe looking only at reserves (I didn’t include Norway—too much of an oil-exporter—so I didn’t really need to adjust for official non-reserve assets). And here is a chart adding the reserve growth of the major oil exporting regions to Asian official asset growth.*** It clearly shows that the growth in official assets was correlated with the big run-up in their combined surplus prior to the crisis (for much more, see Joe Gagnon's 2013 working paper, and his forthcoming book with Fred Bergsten) and that the countries that historically have accounted for most of the reserve build-up are now running down their stock of assets. Broadly speaking, over the last ten years, Asia's surplus hasn't changed much while Europe has replaced the oil exporters as the second main driver of global payments imbalances. And private outflows rather than official outflows have become the financial counterpart to the world's big current account surpluses. That in turn matters for the composition of inflows into the United States: the world is buying fewer Treasuries and far more U.S. corporate bonds—though Asia also seems to have regained confidence in Freddie and Fannie. Taiwanese life insurers, Korean pension funds, and Japanese banks have more risk tolerance than traditional central bank reserve managers. The same is true of German insurers, Dutch pension funds, and for that matter Switzerland's reserve managers—who have more freedom than most of their counterparts to buy equities and corporate bonds alongside traditional reserve assets. Such changes in the composition of inflows to the U.S. could help explain why the IMF found that relative to fundamentals, U.S. corporate spreads seem a bit tight… One last point: The U.S. Treasury suspects that this fall off in reserve growth is largely a function of the dollar's strength, and has expressed concern that it may not be durable in its most recent foreign exchange report. I generally agree: the world's central banks historically have intervened far more when the dollar is weak than when the dollar is strong. The major surplus countries in a sense only need to step up and the use their government balance sheets when the market doesn't want to fund the U.S. external deficit. But I also suspect the world's big surplus countries are also now a bit more skilled than in the past at disguising their intervention—sovereign wealth funds can keep foreign assets off the books of the central bank, government pension funds often do some of the heavy lifting, and, in China's case, the growth in the overseas lending of China's state banks is likely to have structurally reduced the pace of reserve growth in good times. When intervention returns, it may not be primarily through the use of central bank balance sheets. *The long-standing joke that the IMF stands for “It is mostly fiscal” applies to the WEO data set. Tons of fiscal numbers. And for the balance of payments, only the current account (at least for individual countries). Sad! ** The overall U.S. external deficit is of course smaller, thanks mostly to a large fall in the oil deficit. The rise in the offshore income of U.S. multinationals also helped, as did the fall in the rate the U.S. paid on its external borrowing. *** I was lazy and did not try to adjust for the buildup of sovereign wealth fund assets in the Middle East. I have another technical complaint here. The IMF used to provide an estimate of official outflows from the Middle East and North Africa, which was a good proxy for the activity of the region's sovereign wealth funds—it was one of the bits of the WEO data tables that I found most useful. With the shift to the new IMF balance of payments standard (BPM 6) the IMF now only reports aggregated portfolio outflows. The old breakdown was in some ways more useful. The IMF could help to address this by insisting that more countries report the disaggregated data on portfolio outflows—even when this shows that the public sector accounts for the bulk of outflows. **** Singapore's headline reserves aren’t moving, but Singapore’s forward book is now rising and government deposits abroad have soared—call me suspicious.

Most analysis of China’s economy emphasizes the risks posed by China’s high level of investment, and the associated rise in corporate debt. Investment is an unusually large share of China’s economy. That high level of investment is sustained by a very rapid growth in credit, and an ever-growing stock of internal debt. Corporate borrowing in particular has increased relative to GDP. Not all this investment will generate a positive return, leaving legacy losses that someone will have to bear. Rapid credit growth has been a fairly reliable indicator of banking trouble. China is unlikely to be different. Concern about the excesses from China’s investment boom permeate the IMF’s latest assessment of China, loom large in the BIS’s work, and the blogosphere. Gabriel Wildau of the Financial Times: "Global watchdogs including the International Monetary Fund and the Bank for International Settlements (not to mention this blog) have become increasingly shrill in their warnings that China’s rising debt load poses global risks." Yet I have to confess that defining China’s primary macroeconomic challenge entirely as "too much debt financing too much investment" makes me a bit uncomfortable. Investment is a component of aggregate demand. Arguing that China invests too much comes close to implying that, as a result of its credit boom/ bubble, China is providing too much demand to its own economy, and, as a result, too much demand for the global economy. That doesn’t seem entirely right. China’s banks have not needed to borrow from the rest of the world to support the rapid growth of domestic credit. China’s enormous loan growth, counting the growth in shadow lending, has been self-financed; deposits and shadow deposits seem to exceed loans and shadow loans.* Most countries in the midst of credit booms run sizable external deficits. China, by contrast, still runs a meaningful current account surplus. China is exporting savings even as it invests close to 45 percent of its GDP. And even with an extraordinary high level of domestic investment, China’s economy still, on net, relies on demand from the rest of the world to operate at full capacity. That is what differentiates China from most countries that experience a credit and investment boom. An alternative frame would start with the argument that China saves too much. A high level of national savings—national savings has been close to 50 percent of GDP for the last ten years, and was 48 percent of GDP in 2015, according to the IMF (WEO data)—creates an on-going risk that China will either over-supply savings to its own economy, leading to domestic excesses, or to the world, adding to the risks from global payments imbalances. From this point of view, the high level of investment, and the risks that come from high levels of investment, flow in part from the set of policies that have given rise to extraordinarily high levels of domestic savings. After the global financial crisis, the vast bulk of Chinese savings now is invested, no doubt rather inefficiently, at home. Bai, Hsieh, and Song’s excellent Brookings Paper on Economic Activity emphasizes that the surge in investment after the crisis was very much a product of government policy. But even with a high level of investment spurred by rapid growth in domestic credit some Chinese savings still bleeds out into the world economy. And China’s savings exports—exporting savings is an alternative way of describing a current account surplus—create difficulties when most advanced economies themselves are struggling with too much savings of their own, and have trouble putting all the savings now available in their economies to good use. That is what low global interest rates and weak global demand growth are telling us. Thus, from the rest of the world’s point of view, a fall in investment in China on its own poses a set of risks. Less investment means less demand for imports. The imported component of investment is, for now, much higher than the imported component of consumption. China’s recent import growth has been quite weak. It is increasingly clear that the slowdown in Chinese investment in 2014 and 2015 had a larger global impact—counting the second-order impact on commodity prices and investment in commodity production—than was initially expected.** If less investment leads to a shortfall in growth in China and monetary easing, it would also tend to push China’s exchange rate down—resulting in the risk that China would both import less and export more. That isn’t good for a world short on demand and short on growth. From the point of view of the rest of the world, the “win” stems from a fall in Chinese savings, not a fall in investment. Lower savings would mean China could invest less at home without the need to export savings to the rest of the world. Lower savings implies higher levels of consumption, whether private or public, and more domestic demand. Lower savings would tend to put upward pressure on interest rates, and thus reduce demand for credit. Higher interest rates would tend to discourage capital outflows and support China’s exchange rate. That’s all good for China and good for the world. It would result in lower domestic risks and lower external risks. So I worry a bit when policy advice for China focuses primarily on reducing investment, without an equal emphasis on the policies to reduce Chinese savings. To take one example, the IMF’s last Article IV focused heavily on the need to slow credit growth and reduce the amount of funding available for investment, and argued that China should not juice credit to meet an artificial growth target. I agree with both pieces of the IMF’s advice. But I also am not sure that it is enough to just slow credit. I would have liked to see a parallel emphasis on a set of policies that would help to lower China’s high national saving rate. The IMF’s long-run forecast assumes that China’s demographics—and the policy changes already in train (a half point projected increase in public health spending, for example)—will be enough to bring down China’s savings (as a share of GDP) at a faster clip than Chinese investment falls (as a share of GDP); see paragraph 25 of this paper. Even as the off-balance sheet deficit falls and the on-budget fiscal deficit remains roughly constant.*** Mechanically, that is how the IMF can forecast a fall in the current account deficit alongside a fall in investment and a fall in China’s augmented fiscal deficit. So the IMF’s external forecast in effect makes a big bet on the argument that Chinese savings is poised to fall significantly even without major new policy reforms in China. The actual fall in savings from 2011 to 2015 was rather modest, so the IMF is projecting a bit of a change. The BIS also has long emphasized the risks from China’s rapid credit growth. Fair enough: the BIS has a mandate that focuses on financial stability, and there is no doubt that China’s very rapid pace of credit growth is adding to range of domestic financial fragilities. To my knowledge, though, the BIS hasn’t warned that in a high savings economy, slower credit growth without parallel reforms to reduce the savings rate runs a substantial risk of leading to a rise in savings exports, and a return to large current account surpluses. From 2005 to 2007, China held credit growth down through a host of policies—high reserve requirements and tight lending curbs on the formal banking system, and limited tolerance of shadow finance. The result? Fewer domestic risks no doubt. But also a policy constellation that led to 10 percent of GDP current account surpluses in China.**** Those surpluses, and the offsetting current account deficits in places like the U.S. and Spain, weren’t healthy for the global economy. Do not get me wrong. It would be far healthier for China if it didn’t need to rely so heavily on rapid credit growth to keep investment and demand up. China’s banks already have a ton of bad loans and many almost certainly need a substantial capital injection. More lending likely means more bad loans. The risks here are real. But I also would be more comfortable if the global policy agenda put somewhat more focus on the risks from high Chinese savings—as in China’s case, high domestic savings are a root cause of a lot of the domestic excesses. I am not convinced that China’s national savings rate will head down on its own, without any policy help. * See, among others, Tao Wang of UBS—who has pulled together the relevant data in her market research. ** Both the IMF and the ECB have argued that the fall in investment explains much of its recent weakness in Chinese import growth, and thus help explain the recent weakness in global trade. The IMF and ECB papers build on work first done by Bussiere, Callegari, Ghironi, Sestieri, and Yamano. Both Chapter 2 (on trade) and Chapter 4 (on spillovers from China) of the most recent WEO imply that the 2014-15 investment slowdown had larger than initially expected global spillover. *** A technical point. A large government deficit usually lowers national savings. So from a savings and investment point of view, a traditional government deficit tends to impact the current account by lowering savings. But it seems like much of the augmented fiscal deficit—the IMF’s term for the borrowing of local government investment vehicles and the like that doesn’t show up in formal definitions of even the “general government” fiscal deficit—has shown up as a rise in investment. The IMF’s adjustment thus implies private investment (and private credit growth) has been overstated a bit, and public investment understated. So if Bai, Hsieh, and Song are right, a fall in the augmented part of the augmented fiscal deficit would show up as a fall in investment, not a fall in national savings. The line between the state and firms is especially blurry in China, as many firms are owned by the state—but expanding the perimeter of “fiscal policy” to include various local financing vehicles that could be viewed as state enterprises requires some offsetting adjustments. **** The BIS "credit gap" (data here, and here) was negative when China’s current account surplus was at its peak; that isn’t an accident. China was suppressing lending growth through loan quotas and high bank reserve requirements at the time.

With a bit of technical assistance, I was able to do a better job of quantifying the IMF’s recommended fiscal path for Japan. The IMF wants a 50 to 100 basis point rise in Japan’s consumption tax every year for the foreseeable future, starting in 2017. A 50 basis point rise would result in between 20 and 25 basis points of GDP in structural fiscal consolidation a year (the call for the tax increase is in paragraph 23 of the staff report, and is echoed in the IMF’s working paper). The IMF doesn’t want Japan to continue relying on fiscal stimulus packages, which typically have funds for public investment and the like (paragraph 23). As a result, there is a 60 basis points of GDP consolidation from the roll-off of past stimulus packages (the change in the structural primary balance is in both table 1 on p.38 table 4 on p.41 of the staff report). That implies 80 to 85 basis points of GDP in structural fiscal consolidation. But, in the staff working paper (not formal advice, but it clearly reflects the IMF’s overall recommendations), the preferred policy scenario shows an 80 basis point of GDP increase in temporary transfers and public wages to support the proposed incomes policy (this is in the working paper appendix, in table I.1 on p. 33). Net it all out; the result is basically a neutral stance, not the consolidation I initially suspected. The 0.5 percent of GDP fall in general government net lending/borrowing in table 2 on p. 25 of the working paper stems from a fall in interest payments and an increase in nominal GDP that is projected from the new incomes policy.* Actually if you look at table 4 in the staff report, Japan’s is expected to receive more in interest income than in pays out in interest in 2017. Japan’s government is projected receive 1.6 percent of GDP in interest on its assets (including its foreign reserves, which are largely held by the ministry of finance) and pay 1.3 percent of GDP in interest on its debt. The total fiscal deficit is thus smaller than the primary fiscal deficit in 2017. Welcome to the world of negative interest rates. The same methodology can be applied to deduce the IMF’s proposed fiscal stance for 2018. It generates a small proposed fiscal expansion in 2018, as the contraction from the roll-off of past stimulus is smaller (table 3 of the staff report shows a smaller change in the primary balance in 2018 than in 2017) and there is an ongoing increase in temporary transfers. Alas, these calculations are not totally straight-forward: getting the right answer requires two documents, and an appendix table! Using the change in the structural balance rather than the change in the structural primary balance also generates a slightly different result (a modest consolidation in 2017). A box that pulled together and quantified the Fund’s proposed offsets to the consolidation measures proposed for 2017 would have been helpful. I trust numbers in tables more than words. There have been too many stimulus packages in Japan that haven’t actually stimulated the economy, as one package just offset the roll-off of an earlier package.* A flat 2017, slightly expansionary 2018, and then a period of steady consolidation doesn’t actually detract from the core of my initial argument. I was focused on the impact of the IMF’s policy recommendations on global current account adjustment. The Fund is still recommending a medium-term fiscal consolidation in Japan, one that on its own would be expected to raise Japan’s current account surplus over time. And the proposed medium-term fiscal consolidations in the eurozone, Japan, and China still dwarf the (very modest) medium-term fiscal expansions proposed for Sweden and Korea. As a result external rebalancing—a reduction in the size of both external surpluses and external deficits—necessarily would have to be achieved through a larger fiscal consolidation in countries around the world that now run external deficits than in the countries that now have external surpluses. For a fiscal expansion in the surplus countries to contribute meaningfully to external rebalancing, the Fund will either need to recommend sustained fiscal deficits in countries like Japan, or encourage countries with both trade and fiscal surpluses (Korea and Germany) to move into fiscal deficit, not stop at fiscal balance. Such is the math. Paul Krugman though was more irked by the near-term impact of the IMF’s recommendations on demand. I suspect Dr. Krugman would prefer that the Fund embrace a bolder fiscal relaunch, especially with Japan stuck at the zero ten year bound (see Cecchetti and Schoenholtz, among others). A government that makes more on its lending than in pays on its borrowing probably has a bit of fiscal space.** But directionally, in Japan, the IMF is pushing (softly) in the right direction (at least in 2018). And, well, if the IMF can propose a few basis points of fiscal expansion in Japan over the next couple of years, I would hope it could do the same for the eurozone. *** * The paragraph in the staff report that I thought outlined the IMF’s recommended pace of consolidation (paragraph 28) only applies if Japan doesn’t adopt the IMF’s proposed incomes policy. I confess I missed the text in italics that limited the advice in paragraphs 27 and 28 to the "no reforms, no incomes policy" case; mea culpa. ** On net, Japan’s government is forecast by the IMF to receive more interest income than it pays (remember Japan has substantial financial assets, net debt is much smaller than gross debt) from 2017 to 2020 (see Table 3 on p. 40 of the staff report). Perhaps that will change with the BoJ’s new policy, as the Bank of Japan now doesn’t want the ten year JGB rate to be substantially negative—but probably it will not change by much. *** And do so without relying on the creation of new fiscal capacity at the euro-level. Common fiscal institutions are a good idea for the medium-term but the reality right now is that the eurozone’s aggregate fiscal policy is the sum of the national fiscal policies of its member states.

In theory, the IMF now wants current account surplus countries to rely more heavily on fiscal stimulus and less on monetary stimulus. This shift makes sense in a world marked by low interest rates, the risk that surplus countries will export liquidity traps to deficit economies, and concerns about contagious secular stagnation. Fiscal expansion tends to lower the surplus of surplus countries and regions, while monetary expansion tends to increase external surpluses. And large external surpluses should be a concern in a world where imbalances in goods trade are once again quite large—though the goods surpluses now being chalked up in many Asian countries are partially offset by hard-to-track deficits in “intangibles” (to use an old term), notably China’s ongoing deficit in investment income and its ever-rising and ever-harder-to-track deficit in tourism. In practice, though, the Fund seems to be having trouble actually advocating fiscal expansion in any major economy with a current account surplus. Best I can tell, the Fund is encouraging fiscal consolidation in China, Japan, and the eurozone. These economies have a combined GDP of close to $30 trillion. The Fund, by contrast, is, perhaps, willing to encourage a tiny bit of fiscal expansion in Sweden (though that isn’t obvious from the 2015 staff report) and in Korea—countries with a combined GDP of $2 trillion.* I previously have noted that the Fund is advocating a 2017 fiscal consolidation for the eurozone, as the consolidation the Fund advocates in France, Italy, and Spain would overwhelm the modest fiscal expansion the Fund proposed in the Netherlands (The IMF is recommending that Germany stay on the fiscal sidelines in 2017). The same seems to be true in East Asia’s main surplus economies. Take the Fund’s advice on China. The Fund thinks that the right measure of China’s fiscal position is what the Fund calls the "augmented fiscal balance." And the Fund thinks that the augmented deficit is too big, and China should do modest consolidation in 2017.** "The projected increase in the augmented deficit in 2016 is thus not warranted from a structural perspective. Neither is it warranted from a cyclical perspective given the growth outlook. For 2017, a moderate reduction of the augmented deficit seems appropriate. Only if growth threatens to fall sharply (well below staff’s projection of 6.2 percent) should the deficit widen.” (emphasis added; paragraph 36, on p. 23 of the staff report) To be fair, the Fund isn’t calling for on-balance sheet fiscal consolidation, and even seems open to breaching the 3 percent of GDP limit for the headline deficit if that is needed to support more aggressive reforms. Directionally, though, the Fund still wants consolidation. Take the Fund’s advice on Japan. The first consumption take hike—from 5 to 8 percent—didn’t go that well. Consumption never recovered, and the economy lost momentum. But rather than reconsider consumption tax based consolidation, the Fund wants Japan to double down and commit to raise the consumption tax to 15 percent (rather than 10 percent): “a gradual increase in the consumption tax towards at least 15 percent, e.g., in increments of 0.5–1 percentage points over regular intervals, would better balance the objectives of supporting growth and achieving fiscal sustainability in the long run. … Starting the increases as soon as possible and replacing the currently planned 2019 hike with such a pre-announced, gradual path would enhance the credibility of the long-run fiscal adjustment effort..." (paragraph 23, on p. 14 of the staff report) That isn’t exactly a call to use the fiscal arrow to relaunch Japanese demand growth. The Fund is calling for slower and more predictable pace of consolidation, but it is still calling for consolidation, and a form of consolidation that would weigh most heavily on households. Monetary and fiscal policies would work in opposite directions. Japan is a hard case. It has an unusually high level of public debt. It also has an unusually low interest rate on that debt. And fiscal risks are reduced so long as the stock of debt actually held by the public is falling fast: Think of a 5 percent of GDP fiscal deficit and annual purchases by the Bank of Japan (BoJ) of 15 percent of GDP (Or take a look at Figure 3 in Toby Nangle’s paper). With fiscal consolidation at a steady 0.5 percent of GDP pace needed for the next ten to fifteen-plus years according to the Fund (see paragraph 23 of the staff report) it is a little hard to see how the BoJ can stop buying any time soon—so if Japan adopted the Fund’s advice, the fall in debt held in the market would likely be quite fast.*** And what of Korea? The details of the Fund’s advice for Korea are still not out. But there is no real evidence the Fund wants a significant, sustained fiscal loosening in Korea, even though Korea has low government debt, no fiscal deficit to speak of and a $100 billion-plus current account surplus (7-8 percent of Korea’s GDP). The Fund’s forecast, if I read it correctly, projects that Korea’s general government will return to a fiscal surplus of 1.0 percent of GDP over time.**** Korea needs to ease just to avoid tightening, so to speak. Asia’s two big surplus countries (China and Japan) no doubt present a slightly more complicated case than the eurozone taken as a whole. The eurozone is in primary balance, and in aggregate, has only a modest deficit. The absolute size of China’s augmented deficit and Japan’s primary deficit no doubt bother many at the Fund. On the other hand both China and Japan run significant external surpluses even with sizable fiscal deficits; the obvious implication is that without those fiscal deficits the size of their external surpluses would be much bigger. Korea has no fiscal deficit to speak of, and a current account surpluses that is bigger, relative to its GDP, than the current account of either China or Japan. Bottom line: if the Fund wants fiscal expansion in surplus countries to drive external rebalancing and reduce current account surpluses, it actually has to be willing to encourage major countries with large external surpluses to do fiscal expansion. Finding limited fiscal space in Sweden and perhaps Korea won’t do the trick. 20 or 30 basis points of fiscal expansion in small economies won’t move the global needle. Not if China, Japan, and the eurozone all lack fiscal space and all need to consolidate over time. [*] See table 3 of this report for the Fund’s assessment of the gap between a country’s 2015 fiscal balance and its appropriate long-term balance. Sweden should go from a surplus of 0.3 percent of GDP to balance, and Korea from a surplus of 0.6 percent of GDP to rough balance (a deficit of 0.1 percent of GDP) [**] The IMF is not calling, thankfully, for China to reduce its on-budget fiscal deficit, which is forecast to remain at close to 3 percent of GDP for some time. But the Fund is clearly calling for a significant overall consolidation, along with slower growth in private credit. [***] I assume the Fund expects that Japan’s supplementary budget will offset the 0.8 percent of GDP fiscal consolidation that the Fund projects for calendar 2017 in Japan. The staff report didn’t seem to warn of the demand side impact of allowing past stimulus packages to roll-off without being replaced with new stimulus. The tendency toward consolidation from the roll-off of past stimulus packages makes it hard to assess the actual fiscal impetus in Japan, as Japan is on a bit of a treadmill where ongoing temporary stimulus has to be replaced by new stimulus just to avoid a negative fiscal impulse. [****] Korea’s national pension system collects significantly more than it takes in, generating a substantial surplus. The headline fiscal deficit of the government thus differs from the general government’s fiscal balance. Note: I initially wrote that the IMF forecasts Korea’s general government balance to rise to a surplus of 1.5%. It looks to be 1% (using general government net borrowing/ lending). I also corrected a couple of spelling errors.

The eurozone collectively has a substantial external surplus, and its economy is operating below potential. In the framework set out in the IMF’s external balance assessment, that pretty clearly calls for fiscal expansion: "Surplus countries that have domestic slack need to rely more on fiscal policy easing, which would address both their output gaps and their external gaps... Meanwhile, deficit countries should actively use monetary policy, where available, to close both internal and external gaps." But is the IMF following its own advice (for a currency union that has an external surplus and domestic slack, I am well aware of the fact that the eurozone is not a single country with a single fiscal policy) and actually recommending a fiscal expansion in the eurozone? Best I can tell, no. Not for 2017. The IMF of course is for more fiscal stimulus at the European level. But that is a hope, not a reality. The capacity for a common eurozone fiscal policy conducted through borrowing by the center doesn’t currently exist, and it realistically isn’t going to materialize next year. That means the eurozone’s aggregate fiscal impulse is the sum of the fiscal impulses of each of its main economies. What does the IMF recommend there? In Italy the IMF seems to want about a half a point of structural fiscal consolidation (see paragraph 35 of the staff report). In France, the same (see paragraph 33 of the staff report). In Spain, the last IMF article IV called for a half a point of consolidation as well (see paragraph 33). That though is likely to be ratcheted up, as Spain hasn’t done much consolidation over the past two years and now will need to engage in a major consolidation to get to the Commission’s 3 percent of GDP fiscal target in 2018. Spain, Italy, and France collectively account for a bit more of the eurozone’s GDP than Germany and the Netherlands. So, is the IMF recommending enough fiscal expansion in Germany and the Netherlands to assure a positive fiscal impulse for the eurozone as a whole? No. The IMF is now advocating that Germany avoid returning to a structural surplus, and invest any windfall savings from higher than expected revenues or lower interest rates. But the IMF isn’t advocating Germany do any more on net than it already had done. Germany delivered a significant stimulus is 2016, but it was a one and done stimulus. Germany isn’t projected to generate a positive fiscal impulse in 2017.* The Netherlands needs stimulus on purely domestic grounds given persistent weakness in household demand and an ongoing output gap (see the "Tricky Balance" chart in the IMF’s external balance assessment). And the IMF did recommend a modest fiscal expansion in its last assessment of the Dutch economy. But also it didn’t protest too much when the Dutch government politely declined. See paragraphs 14 and 15 of the 2015 staff report. Bottom line: Germany’s 2017 fiscal impulse is projected to be neutral. The fiscal expansion the IMF recommended in the Netherlands for 2017 is not likely to be adopted and even if adopted it would be too small to offset the fiscal consolidation the IMF is recommending in Spain, Italy, and France. Given the weight of France, Italy and Spain in the eurozone, this implies the IMF is recommending that the eurozone as a whole consolidate next year. And in the IMF’s global framework, that implies the IMF is currently recommending policies that would tend to raise the eurozone’s external surplus. That goes to a larger global problem. The IMF’s framework for looking at external balances says, more or less, that intervention in the foreign exchange market has only a modest impact on external balances.** The policies that matter are healthcare spending, and of course, fiscal policy. The IMF always lives up to its mostly fiscal reputation. The implication, if the IMF wants to be symmetric and worry about global demand as well as global balance of payments adjustment, is that the IMF needs to be as aggressive in recommending fiscal expansion in surplus countries as it is in recommending fiscal consolidation in deficit countries. Judging by its recommendations in Europe, it still has a way to go. In addition to watching the IMF’s aggregate recommendation for the eurozone, I also will be watching the IMF’s 2017 fiscal recommendation for Korea—which has a German-sized current account surplus and, broadly speaking, a German fiscal policy (counting the off-budget surplus in Korea’s social security fund, Korea ran a general government surplus in 2015).*** And its last stimulus package seems to lack what the Japanese would call fresh water. There is of course a second, more straight forward argument for why the IMF might want to encourage Germany to do a bit more public investment in 2017: It offers the most obvious way to help insulate Germany from a slowdown in British demand. The IMF’s initial analysis of Brexit suggested it might knock 20 basis points off eurozone growth in 2017. The hit to Germany, though, was bigger—more like 40 basis points. Logical, given Germany’s greater dependence on exports. Germany’s 2017 nominal wage growth—judging from the contracts agreed this spring—also looks to be slower than in 2016. All in all, it would seem like there is ample reason for the IMF to encourage Germany to offset the impact of the drag expected from Brexit with a looser fiscal policy in 2017 and 2018, even if that means Germany would need to run small structural fiscal deficits for a time. It would support German growth in the face of an expected external shock. It would help ease the pressure on monetary policy when the ECB is at the zero bound. It would make it easier for Italy, Spain, and France to offset the demand drag from fiscal consolidation through exports, and thus facilitate the eurozone’s internal rebalancing. And it would help reduce the eurozone’s contribution to global current account imbalances. Note: edited slightly after publication; I had "internal" when I meant "external" in the last paragraph * Germany did provide a solid (around 1 percent of GDP) fiscal stimulus in 2016 in response to the migrant crisis, as the IMF estimates it moved from a 70 basis points of GDP general government surplus to a 30 basis point of GDP general government deficit. But now that Germany no longer has a surplus, the politics of further stimulus get complicated. The IMF’s staff report emphasized that any further public investment needs to be Within the fiscal rules ("Fiscal resources available within the envelope defined by the fiscal rules, including in case of overperformance, should be used to finance additional public investments...") and the debt brake doesn’t leave much space (see paragraph 49 of the 2016 staff report; the fiscal impulse is from table 2 of the staff report. ** This gets technical. In the IMF’s analytical framework, the intervention variable is interacted with the capital controls variable, reducing its impact. The capital controls variable is never “1.” For China, for example, it was 0.5 last year. So a percentage point of GDP in Chinese intervention could not have more than a roughly 17 basis point impact on the current account (the 0.45 intervention coefficient is multiplied by capital controls, which are now rated at 0.38 on a 0 to 1 scale. For Korea, a percentage point of GDP intervention could not have an impact of more than 6 basis points, given the 0.13 intensity of Korea’s controls. This is well below the impact that the Peterson Institute’s Joe Gagnon has estimated. And the impact is further reduced by the fact that the relevant variable is the gap between a country’s current level of controls and the optimal level of controls. So if the IMF (wisely) believes China isn’t ready yet to liberalize its financial account, the impact of intervention in generating a “policy gap” falls further. Right now, given the size of the capital controls gap for China, a dollar of intervention in China results in only a ten cent change in the current account. It doesn’t matter much now, as emerging markets sold reserves in 2015 and so far in 2016. But the net effect is that in the IMF’s framework, intervention never had much of an aggregate impact even back when there was a lot of intervention. *** The IMF is recommending medium-term fiscal consolidation in Japan, China, and the eurozone. Fairly substantial consolidations too (see table 3 of this report, look at what the IMF calls P* or "optimal" policies and compare that with the current estimated fiscal balance; the full underlying analytics are here). It is recommending a small medium-term fiscal expansion in Sweden and Korea. Both countries ran general government surpluses in 2015, and the IMF is calling them to bring those surpluses to zero or run a small (10 basis points of GDP) general government fiscal deficit. But the resulting swing in the fiscal balance is modest relative to the size of the Swedish and Korean current account surplus. For Germany, the IMF is recommending a fiscal expansion relative to its 2015 surplus; but, as noted above, that stimulus occurred in 2016, and the IMF isn’t currently calling for more stimulus. I should note here that the recommended fiscal stance in the IMF’s external balance assessment is for the medium term; just because the Fund recommends a medium term on-budget consolidation for say China doesn’t mean it recommends an immediate on-budget fiscal consolidation. That said, given China’s extremely high national savings rate and the need for corporate deleveraging, I personally do not see why China could not sustain on-budget fiscal deficits of around 2 percent of GDP in the medium term.