Dollar Regime Change and Southeast Asia - Who Benefits?

During the last 18 months, for most Asian economies, the curtailing of cross-border financial flows associated with the rising dollar has trumped domestic cashflows in setting overall monetary and financial conditions. Now the immediate pressure on cross-border financing is abating, it’s time to look again at Asia’s domestic cashflows.

Private sector savings surpluses are key indicators in identifying the size and direction of the flow of cash between the private sector and the financial system, and so in turn also are important components in determining movements in bond yields.  The more developed the financial system, the more buffers are in place to mute the impact of changes in private sector savings habits. So let's look at the balances for Southeast Asia and India.

Conclusions? Whilst in Indonesia and Malaysia, domestic trends are pushing the private sector savings balance towards inflection points which international finance are likely to notice and react to, the broader beneficiaries of a general relaxation in cross-border financing conditions are likely to be found precisely in those economies where competing pressures are not currently felt, which means the Philippines and India.  Meanwhile, in Singapore and Thailand, the huge savings surpluses already maintained means changes in international financial conditions are unlikely to be noticed greatly.

Not every Southeast Asian economy has a similar savings surplus/deficit profile or trajectory, so the opportunities raised by the weaker dollar and relaxation of cross-border finance flows are not shared equally.

Indonesia and Malaysia are the economies most obviously exposed to the change in the international financial environment, since both countries are approaching the inflection points in the balance of private sector savings and deficits.  For Indonesia, it seems possible that the improvement seen in 1Q’s trade position will allow the private sector to break into savings surplus for the first time since 2012. Indonesia has been running a savings deficit since 2012, which peaked at 2.7% in mid-2013, but which had recovered to a deficit of just 0.7% of GDP by end-2015 and may well have broken into surplus during 1Q16.

It has been a slow and frustrating process getting to this inflection point. But if it occurs, it will mean that Indonesia’s private sector is no longer fundamentally seeking cashflow from the financial sector in order to maintain its current levels of consumption and investment, but rather that those consumption and investment choices are sufficiently modest to allow for net savings/profits which the financial sector needs to redeploy into public sector or foreign assets.  This domestically-generated financial tailwind is likely to be strengthened by any relaxation in cross-border financing conditions.

Malaysia potentially represents the opposite situation: the country has been gradually working its way through a savings surplus which peaked at 16.6% of GDP in mid-2009 to just 2.7% in 2015.  Stabilization of this negative trend has seemed possible on three occasions, but in each case the negative trend has resumed. What a relaxation of international financing conditions would offer Malaysia is a softer landing if the private sector does eventually lapse into savings deficit.

By contrast, Singapore and Thailand already run such huge private sector savings surpluses that their domestic financial conditions are unlikely to be constrained by internal cashflow issues in the first place. So for both Singapore and Thailand the relaxation in cross-border financing conditions is unlikely to have a significant impact on domestic economic or financial activity. Singapore has run a massive savings surplus for as long as anyone has been counting, and by end-2015 it was running at 18.8% of GDP.  Thailand, meanwhile, has recovered from a marginal deficit in mid-2013 to a surplus equivalent to 11.1% of GDP by end-2015. This dramatic build-up really took off  after the military coup of mid-2014, suggesting it is the result of a rise in precautionary saving responding to political uncertainty.  If so, the savings surplus is likely to be maintained, and act primarily as a damper on domestic demand.

What binds Singapore, Thailand, Indonesia and Malaysia together in this analysis is that domestic private savings surpluses or deficits trends are already in place which will tend to be the decisive influence, with a significant relaxation of cross-border financing availability offering only to qualify those conditions, rather than to supersede them.  Counter-intuitively, it is those economies in which currently the private sector savings surplus/deficit situation or trend puts no clear pressure on domestic financial activity, or offers no obvious new opportunity, that the impact of improved cross-border financing conditions is likely to have a more decisive directional influence. Two economies which come into this category are the Philippines and (outside Southeast Asia) India. 

Philippines has a surplus of 3.8%: this ratio has been falling gently since 2013, but may now be stabilizing or even growing once again.   This week, India announced its 4Q current account balance showed a deficit of US$7.07bn or 1.1% of GDP (the trade deficit narrowed US$3.4bn qoq to $34bn, whilst the surplus of invisibles rose US$0.2bn to $18.1bn), which in turn suggests India is running a private sector savings surplus of a steady 2.5% of GDP.  In neither case does movement of the ratio present any immediate difficulties or offer any significant opportunities; in both cases, therefore,  the absence of such immediate prompts means any improvement in international financing conditions are likely to be felt more obviously. 

US 4Q: Profits Fall as Wages and Dividends Rise

The third revision of US 4Q GDP growth arrives garnished with details of how national income was shared between workers, companies and shareholders. The message emerging is that even though growth is slowing, the tightening labour market means employees are managing to secure a slightly greater proportion of the marginal growth.  But in addition, as the overall profits picture also deteriorates, so companies are prepared to pay out a higher proportion of their profits in order to keep their shareholders happy.  This combination is hardly sustainable, and adds more detail to the internal dynamics of the US's profits recession.

US 4Q GDP grew by an annualized 1.4%, according to the third (and final) estimate, double the rate initially estimated, and better than consensus expected.  Growth of private consumption was upgraded to 2.4% from an initial 2.2%, thanks to a 2.8% jump in services. Residential investment also grew by an annualized 10.1% (rather than the 8.1% initial thought), and net exports stripped only 0.14pps from GDP growth, rather than the 0.47pps initially estimated.

But not everything was upgraded: private non-residential investment spending fell an annualized 2.1%, worse than the 1.8% fall initially estimated.

The 1.4% annualized GDP growth just about keeps the economy in touch with its 2009-2015 trend growth rate of 2.1%, but there’s no hint that the economy is moving back towards its 1990-2007 trend growth rate of 3.2%.

More worryingly, despite the upgrade, this GDP expansion is not enough to sustain profits growth. In fact, corporate profits, including depreciation and inventory adjustments, fell 9.2% qoq, and generated the steepest yoy fall since the recession of 2008. Proprietor’s income fared slightly better, rising 0.5% qoq, which suggests a higher proportion of profits are being paid out in dividends. Rental income, meanwhile, rose 1.2% qoq.  These three sources of income from profits together fell 3.7% qoq and fell 3.4% yoy, which, once again, was the bleakest result since 2008.

By contrast, the wages bill rose 1% qoq and 4.1% yoy.  The result is that the wage bill is now rising faster than GDP  in nominal terms, and wages as a proportion of GDP rose 40bps yoy to 54%, which was the highest proportion since 2009, and almost exactly in line with the 10yr average. Meanwhile, corporate profits as as percentage of GDP fell 1.4pps yoy to 8.3% in 4Q, which was 0.4 standard deviations below the 10yr average of 8.8%. If one includes proprietors’ income and rental receipts, the proportion rises to 19.8% of GDP,  also down 1.4pps yoy, but still above the 18.7% average of the last 10 years. 

Why US Capital Goods Orders Keep Sinking

Although February's US capital goods numbers were routinely dreadful, in some ways they were not quite as bad as they first seemed, since they implied no significant disequilibrium within the capital goods sector itself. Nevertheless, with return on capital drifting lower, we should not expect any sustained improvement in the short and probably medium term: most likely orders are going to keep falling.

February's numbers were bad enough: capital goods orders (nondef ex-air) fell 1.8% mom, to the lowest dollar value since December 2013, and  shipments fell 1.1%, to the lowest dollar value since January 2014,  Hardest hit were orders for electrical equipment (down 2.8%) and machinery (down 2.6%); for shipments  electrical equipment fell 2% and machinery fell 1.3%..

It’s not exactly good news, but it does perhaps mitigate the shock: despite the falls, there was no substantial deterioration in the balance between orders, shipments and inventories.  Shipments fell 1.1% mom and inventories fell 0.3% mom, and that was enough to push up the inventory/shipment ratio by only 0.01pt to 1.76x. Any rise is disappointing, but the 1.76x ratio is actually still below the average since 2010, and implies no irresistible pressure to cut orders and start dumping inventory.  Second, with orders down 1.8% mom and shipments down 1.1%, the book-to-bill ratio settled at almost exactly 1, with a fractional fall well within the margin of error. Once again, it’s not great, but it doesn’t imply that the industry is massively out of equilibrium. And that’s the good news: the capital goods sector is in retreat, but that retreat isn’t developing its own negative feedback cycle yet.

The much poorer news is that the fundamental problem depressing capital spending remains unaddressed: return on capital is almost certainly falling, and until it starts rising again, there’s good reason to expect spending on capital goods to keep falling too. In the chart below, the ROC directional indicator is calculated by expressing nominal GDP as an income from a stock of fixed capital. The stock of fixed capital is estimated by depreciating over 10yrs all non-residential private gross fixed capital formation reported in the quarterly national accounts. 

The chart compares the dollar value of capital goods orders (nondef ex-air) and movements in that ROC directional indicator.  It shows capital orders to be fairly acutely sensitive to movements in the ROC directional indicator.  This has been particularly true since 2010, where even small deflections in the ROC directional indicator have coincided with short-lived rises and falls in capital goods orders.  

As of 4Q15, the stock of capital is growing at 4.1% yoy, whilst nominal GDP is growing only 3% yoy.  It would nominal GDP growth annualizing at 5.1% during 1Q16 just to stabilize this ROC directional indicator.  A gain of that size would surprise consensus.  However, until nominal GDP growth can overhaul the pace of growth of capital stock, the ROC directional indicator will keep drifting down, most probably taking capital goods orders down with it. 

Mario Draghi as (Successful) Football Manager

It is sometime said that the chief skill of a football manager is to pick to which team he should lend his magic at any one time. Much the same  might be said of central bankers and their policy initiatives. For example, when Mark Carney accepted the post of Governor of the Bank of England in late 2012,  it could be inferred that he was fairly confident the boat was not going down.

Something similar may be true of the European Central Bank’s latest clutch of easing measures. The combination of further interest rate cuts (including more deeply negative interest rates on banks’ deposits with the central bank; an enlargement and extension of its asset-buying program, and a new round of even-more generous long-term refinancing offers) probably can’t secure the ECB’s inflation target of near-but-not-quite 2% in the medium term. Nor can they guarantee the rehabilitation of the Eurozone’s banking system: the Japanese experience of the last 25 years shows how hard it is for even the most extreme monetary policy to resuscitate broken banking systems.

However, ECB governor Mario Draghi has pushed the new measures at a time when it is reasonable to expect the Eurozone’s current grinding and acyclical expansion to continue and  accelerate modestly. If so, future commentators will congratulate M Draghi on the part he played in securing it.

In more detail:

i) the ECB’s refinancing rate is cut 5bps to zero; its marginal lending facility rate is cut by 5bps to 0.25%, and the deposit facility rate charged on banks’ excess deposits with ECB is raised 10bps to 0.5%. 

ii) the ECB raised its target for monthly purchases of assets by Eu20bn to Eu80bn a month, with investment euro-denominated corporate bonds included.

iii) a new series of four refinancing operations will be launched in June, each with a four year maturity, under which banks will be able to borrow up to 30% of a specific eligible portion of their loan-book as of end-Jan 2016, with the rate set at the main refinancing rate (ie, zero). 

The Eurozone is still in the early stages of a weak recovery, with 4Q’s 0.3% qoq GDP growth being the 11th successive quarter of sustained fractional growth. This sustained expansion owes nothing to credit growth, but everything to rising employment, which in turn reflects in labour productivity gains wrung out of a labour force in the teeth of a shrinking and aging capital stock.  It is not exciting, but it is durable, and it looks very similar to the post-crisis expansions seen already in the UK and US. 

At the heart of this grinding recovery are improvements in the return on capital, and rising labour productivity. Looking first at return on capital: by 4Q, real GDP was growing at 1.6% yoy, and nominal GDP was growing at 2.6% yoy, whilst gross fixed capital formation was growing at 3.4% yoy in real terms, and 4.1% yoy in nominal terms. Eurozone’s nominal capital stock finally stopped shrinking during the middle of 2015, and by the end of the year was growing at about 0.3% yoy. With capital stock growing only 0.3% whilst nominal GDP is growing around 2.6%, the return on capital must be rising. Moreover, unless nominal GDP growth slumps below 0.3% yoy,  that rise in return on capital will continue for the foreseeable future.  

How is it being achieved?  The answer is: in the same way as we have previously seen in the US and the UK. This recovery is driven by a rise in labour productivity attributable to something other than rises in capital per worker (longer hours? more efficient working? upgraded skills?). As the chart below shows, real output per worker, deflated by changes in capital per worker, has been rising modestly but continuously since 2013, and by the end of last year, output per worker was rising 1.4% yoy. This laid the foundations for a similarly modest but sustained rise in employment. By 4Q15, employment was rising 1.2% yoy, with 4.5mn jobs added since the nadir of 1Q13. 

That slow and steady rise in employment is also responsible for a slow but steady recovery in demand, achieved in the absence of any noticeable credit cycle, and despite the private sector’s unchanged financial caution.  In fact, the private sector is running a savings surplus of just over 5% of GDP, as it has been since 2013. By definition, this savings surplus means the private sector is making more new deposits into the Eurozone banking system than it is taking out in new loans. Nothing the ECB does to interest rates or refinancing opportunities can result rises in net new lending until the private sector waives this financial caution. Even when you flood the banking system with liquidity, as M Draghi is doing, banks’ loan/deposit ratio will fall as the private sector generates net savings flows. 

This continued deleveraging also means that although ECB action may be able, finally, to encourage growth in money M2, the impact of that monetary growth on GDP will be muted by an answering fall in monetary velocity (GDP/M2). What’s happening here is that whilst M2 is a count of the stock of money (essentially cash and deposits), what matters for economic activity in the short term are the activity-creating flows between those accounts.  And the problem is that banks are either unwilling or unable to act effectively to re-circulate those funds around the private economy. 

Nevertheless, even with savings surpluses undiminished, and even with banks’ ability to lend therefore circumscribed, the productivity-based rise in employment is sufficient to allow a steady expansion of demand, even in the absence of the sort of credit-based accelerators usually experienced in a classic business cycle.  This is precisely the sort of steady, acyclical grinding recovery we have been watching for the past few years in the UK and US. 

Three Ways to Track Cross-Border Credit

Summary: Whilst the destruction of wealth represented by the falling stockmarkets will surely have malign economic consequences, they were not themselves responding to problems or excesses in the global economy which would be predictably resolved by a business cycle downturn. Rather, those falls were generated by a contraction of cross-border bank lending which ultimately had its roots in the surge in the dollar during 2Q14.  If so, tracking the inflection in trends of cross-border financing becomes the most important task economists face currently.  Here I present three ways we might be able to get on top of these trends. 

I have argued that whilst the destruction of wealth represented by the falling stockmarkets will surely have malign economic consequences, they were not themselves responding to problems or excesses in the global economy which would be predictably resolved by a business cycle downturn. Rather, those falls were generated by a contraction of cross-border bank lending which ultimately had its roots in the surge in the dollar during 2Q14. The fall in cross-border financing acts on reserve money in just the same way a contraction in central bank lending to the financial system does. It is this unrecognized monetary crunch which toppled stockmarkets, with the worst impacts being felt precisely in those economies (and stockmarkets) which had previously benefitted most from the previous rise in cross-border lending. In practice that meant emerging markets, and China in particular.

The Bank for International Settlements data has the widest record of this, and it shows net cross-border claims on banks dropping by US$748bn to just $496bn between 3Q14 and 3Q15.

The impact was muted for developed country banks, net claims fell by only $185bn to $1.243tr, but for others, net claims dropped $563bn to result in a net liability of $747bn.  This hit to ex-developed country banks was only partly offset by a continued rise (of US$190bn) in net claims on non-bank borrowers in the same period. Overall, however, net claims on countries outside the ‘developed’ definition fell by US$373bn - or by just over half - to US$371bn at the end of 3Q15. 

A couple of things strike me visually concurrent and also what you’d expect: the direction of the flows mirrors the strength and weakness of the dollar/SDR. When the dollar strengthens mid-2011 to mid-2012, and again mid-2014 to early 2015) cross-border claims tend to diminish; when the dollar stabilizes or weakens, cross-borders lending picks up once again. 

Because my global Monetary Conditions Indicator (excluding China) includes direction and volatility measures of movements of the US dollar (as well as monetary aggregates, real interest rates and yield curve measures), it is not surprising that there seems to be a good relationship then, between that and the movement of total international bank net claims on the the world (ex-developed countries). 

The BIS’s data is invaluable in tracking this fundamental stress which finds expression in financial markets, but it is slow arriving - the latest data runs only to September 2015.  To work out the degree to which those stresses are persisting, intensifying or retreating we need faster data. I have three suggestions:  

1. direct observation of foreign assets and liabilities of developed-country bank balance sheets;
2. movements in my global monetary conditions indicator; 
3. activity in credit default swap markets. 

1. Developed Economy Bank’s Foreign Assets/Liabilities

The most direct indicator of the trends in cross border lending are, of course, movements in the net foreign assets or liabilities of developed economy banks. Although obviously less complete than the BIS’s survey, they are updated far more quickly. Looking at these balance sheets, we find two main sources for the drain in cross-border lending: the Eurozone, and, specifically, Hong Kong’s lending to China. 

Eurozone Stress, the Big Unfinished Factor 

As the chart shows, the net foreign asset position of the Eurozone’s banks has been in sharp and almost continuous contraction since the middle of 2014. In July 2014, the Eurozone banking system’s net foreign asset position hit a peak of US$1.917tr, but by January 2016 this had shrunk by US$481bn to US$1.437bn. There is no sign that this contraction has yet come to an end. Moreover, it is a mistake to view this contraction purely as a function of the 19.8% fall in the Euro against the dollar during this period: first, because it is a net figure and the fall in the currency affects both foreign assets and liabilities similarly; second, although Sterling has suffered a similar fall during the same period (down 15.5%), there has been no similar sustained run-down in foreign assets from London’s banks.  Rather, this sustained fall in Eurozone banks’ net foreign assets represents a genuine disentanglement of Eurozone banks from ex-Eurozone finance. The data up to January contains no suggestion that this withdrawal has been completed. Anecdotal evidence and the demonstrated travails of the Eurozone banking system (loan growth of 0.1% yoy in Jan 2016) also give no obvious reason to expect an early reversal.

Hong Kong & China Finance. The second source of stress is the way in which Hong Kong’s financial system has pulled in credit lines to China - a contraction which has been far stronger than the overall fall in its net foreign asset position. I have written about this previously, but to summarise: between June 2014 and NOvember 2015, Hong Kong banks’ net foreign asset position fell US$47bn to US$253bn. This was a relatively modest reversal in trend, but hides a much more extreme reversal in its China position: between June 2014 and November 2015, the banking system’s net China assets fell US$247bn to just US$90bn. I expect by now that net total has already fallen to zero, or possibly less. 

By contrast, the US banking system shows a fall in net foreign liabilities which is the counterparty of the fall in other banking system’s net foreign assets. Between July 2014 and December 2015  US banks’ net foreign liabilities fell by US$338bn to US$1.93tr, as gross foreign assets fell by US$27.6bn, but foreign liabilities fell by $395.6bn. We have data up to the end of 2015, and there is no obvious change in the trend.

Finally, we come to Japan, for which we have the data up to the end of 2015. This shows the familiar run down in net foreign assets between July 2014 and June 2015, although it was a relatively small affair - the contraction was only US$42.9bn. However, uniquely, that was subsequently reversed, and the current US$269bn in net foreign assets held at end-2015 is almost identical to the total held in July 2014. Small though the movement is, this is the only banking system I can find where foreign asset holdings are actually now rising.

2. Global Monetary Conditions Indicator

As we saw in the chart above, inflection points in my global monetary conditions indicator do seem to have coincided with similar inflection points in total cross-border financing flows. It is therefore worth noting that the decline seen between summer 2014 and sustained throughout 2015 does appear at least to have bottomed out - although it is too early to be certain it is inflecting upwards. 

3. Credit Default Swaps - Weekly to end-Feb

The final indicator which may be worth watching is movements in the notional amount of credit default swaps outstanding. The data is timely and delivered weekly from the International Swap Dealers Assn, but has its own difficulties: responding to regulations which demand increased capital accounting for these products, the industry has adopted various ways of ‘compressing’ the notional outstanding balances, essentially by writing down the notional amount of matching or offsetting swaps outstanding in tranches as the cashflows materialize. As this has progressed, since the overall trend has been for the notional outstanding to fall by approximately 17% a year.  

However, when detrended, the movements against trend shows the characteristic movement seen in cross-border lending, and  in particular, the sharp fall against trend since the middle of 2014, which has been sustained until now. 

After a particularly harsh fall at  the end of the year - almost certainly in response to the approach of the end of year balance sheet - there has been something of a recovery in January and February. In fact, the middle of February saw the break above trend since December2014.  Probably not too much should be read into that just yet: it may well just represent a rebound 

infrom the unusually sharp falls of December. Nevertheless, I think this total is worth watching, since a sustained rise vs trend is a plausible early-indicator for a broader recovery in cross-border finance.