UK trade to party like it’s 1992? Watch out cyclical and structural headwinds

This post argues that cyclical and structural factors represent headwinds to the hope that the UK is set to emulate the post-ERM depreciation period and enjoy a net exports bonanza in the wake of sterling’s recent sharp depreciation. The main points are:

  • The post-ERM story is more nuanced than often reported. While UK nominal exports did indeed boom (rising an average of 10% y/y 1993-1997), import growth wasn’t far behind so the trade trade balance only improved by around 1.5pp of GDP. And most of the current account deficit narrowing occurred before sterling’s fall.
  • World trade growth prospects look considerably weaker than in 1992: the WTO expects the weakest world trade growth since the financial crisis, with no high import growth regions anticipated (see Chart). With protectionism on the rise, thats’s a tought environment for the UK to be trying to forge new export markets.
  • Fundemantal structural changes such as offshoring and increased specialisation seem to have reduced the net export benefits of currency depreciations relative to 1992.
  • Previous BoE analysis found that UK net exports disappointed after sterling’s 2007-09 depreciation as UK (services) exporters raised margins and UK suppliers had difficulty subsituting for imports. The weak recent capex environment, likely exacerbated by Brexit uncertainties, raises the chances of such developments repeating.
  • Despite sterling’s recent sharp falls, the real £ERI isn’t yet significantly below long-run averages. And £ undervaluation (overshooting) could be required for UK exports to thrive with weak world demand: demand elasticities exceed price elasticities.
  • While I’ve previously argued that the funding of the UK’s 5.7% of GDP current account deficit has become more fragile, FX valuation effects cause me to aim off Goldman Sachs’ recent FEER-based estimate of £ remaining 10% overvalued. The UK current account deficit improvement will more likely to driven by better foreign investment income than via improving net exports.
  • Despite some recent relative £ stabilisation (markets looking for positives in political statements), near term £ remains vulnerable to widening risk premia as well as being hamstrung by the lowest G10 real interest rates. A further leg-down in early 2017 seems likely as political and economic realities start biting.
  • But with a November MPC rate cut now looking less likely (if not impossible) and potental squeezes of extended £ short positions investors may well shift to selling £ rallies strategies in the near-term with £ consequently range-bound.
  • £/$ could, however, fall below 1.20 on a December FOMC rate rise (around 70% priced) while €/£ would likely fall back should the 4 Dcember Italian referendum produce a “no” vote (the polls are too close to call) or should the ECB extend QE in December.

slide11slide9

Help us net trade contribution, you’re our only hope….but the post ERM experience is more nuanced than often reported

Brexiteers hope that leaving the EU will free the UK to make fresh trade deals with non-EU countries, offsetting the likely trade destruction from probably inferior EU single market access terms (UK government continued hints at cherry-pick a special deal run counter to continued rebuttals from EU governments).  The further informal inference is that sterling’s recent sharp depreciation (which I  predicted prior to the referendum) will  decently boost to UK net trade and hence support growth. Brexiteers tend hope that the UK can emulate the post-1992 ERM departure experience, when UK exports grew strongly and the current account deficit narrowed.

A first problem is that the post-ERM story is more nuanced than often reported: sterling’s depreciation had a limited impact on the current account deficit (most of the adjustment preceded it). UK exports did indeed grow strongly in the following five years: the chart below illustrates that UK nominal export growth averaged around 10% y/y 1993-1997.  But nominal UK import growth wasn’t far behind (as UK domestic demand recovered). So the UK trade balance improvement wasn’t massive: it went from -0.9% of GDP in Q4 1992 to +0.5% of GDP in Q4 1996 (see second chart). And that drove a similar UK current account improvement (given that primary and secondary income were broadly flat), taking it broadly into balance.  Importantly, however, most of the narrowing of the UK current account deficit from its low of close to 5% of GDP occurred before sterling’s post-ERM fall: the 3pp of GDP improvement over the 1989-92 period reflecting the UK recession depressing UK imports.

slide3slide4

The rest of this piece argues that the current situation and prospects are more adverse that in the 1990s, with structural changes adding further headwinds. That’s setting aside the fact that “gravity” trade models don’t advocate substituting trade deals with distant countries for those with your closest neighbours (of course there’s considerable uncertainties about the precise trade relationships which eventually emerge).

The starting point: weak UK net trade but sharp sterling depreciation since November

The starting point for the hoped-for improvement is net exports isn’t particularly auspicious. The chart below illustrates that net exports have tended to drag on growth in past couple of years: they made a -0.8pp contribution to Q2 GDP.  Of course, sterling has depreciated sharply in recent months (as I  predicted prior to the referendum) which should provide some support to net trade. The question is how much. Surveys are a bit mixed on whether near-term UK export prospects have signifucantly improved.  The long-running CBI industrial trends survey isn’t particularly optimistic: the September survey found release reported a small fall-back in export orders despite sterling’s fall. The September PMIs appear a little more positive, if not yet booming. Specifically, the manufacturing PMI reported the highest export orders since January 2014, although sterling’s fall also generated the highest input price sice 2011.  The Services PMI survey reported improved export orders, although orders remained below average. I discuss caveats around such qualitatitive surveys here.

slide11slide8

Importantly,  All the evidence is that export market prospects are a more important determinant of a country’s export performance than exchange rate movements: demand elasicities are typically around unity (a 10% improvement in foreign demand equates to 10% higher export volumes) while price elasticities typically lie in the -0.4 to -0.7 range.

Cyclical headwinds: Weak world trade prospects represent a major challenge

Unfortunately, world trade prospects are looking weak: it’s going to be a tough market for the UK to forge new trade links in. Specifically, the Dutch National Bank CPD meaure of world trade has slowed sharply since 2011 – recording -0.9% y/y in July, down from over 4% in 2014 and (see Chart). That picture is very different from after the 1992 sterling depreciation: world trade was growing at around 10% and peaked at around 14% y/y in 1994. World trade growth also rebounded after the financial crisis (after falling very sharply): growing at nearly 20% y/y in mid-2010.

The recent IMF WEO trade chapter (see here) attributes three-quarters of the world trade disappointment to weak activity, especially the dearth of investment. But the IMF also highlight the waning pace of trade liberalization/uptick in protectionism and the declining growth of global value chains. Bank of England Analysis instead argues that the trade growth slowdown reflects a greater share of world activity being accounted for by countries whose imports grow more slowly relative to GDP.  But none of those explanations look positive for UK export prospects since they are not expected to reverse.

Indeed, the World Trade Organisation (WTO) recently sharply cut their world trade growth forecasts (see here): to only 1.7% in 2016 (from 2.8% previously) and 1.8%-3.1% in 2017 (from 3.6% previously, the significant uncertainties explain why they quote a range for the first time ever). That’s the slowest growth in world trade since the financial crisis. The WTO also highlight that the global trade to GDP growth ratio has recently fallen to an unprecedentedly-low 0.8, rather than the more normal values around 2. In summary, the WTO argued that “The dramatic slowing of trade growth is serious and should serve as a wake-up call. It is particularly concerning in the context of growing anti-globalisation sentiment.”

slide9slide10

And unfortunately the detailed WTO forecasts reveal that there’s not a prospective high-growth region for the UK to try and re-orientate its’s trade flows towards in forging new trade deals after leaving the EU. Specifically, South and Central America is projected to have marginally the fastest 2017 import growth – a mid-point of 2.95%, versus the 2.45% mid-point for the world – but that’s after plummeting in 2016 (-8.3%).  Interestingly, Europe’s forecast import growth is in-line with the global average (2.45% mid-point for 2017), is only marginally below North America and only slighly further below Asia (2.65%). Of course, individual countries will inevitably under/over-perform within that big picture, but the challenge to the UK is finding a booming market to buy it’s exports is nevertheless clear.

Structural factors also argue for caution

Structural factors may also be working against the Brexiteers’ high hopes for a UK net export boom with new trade deals and sterling’s depreciation.

In particular, Bank of Canada Governor Poloz recently argued that trade integration and production offshoring have reduced the trade impacts of depreciations as well as increasing the inflation impacts. MPC themselves raised the persistence of their exchange rate pass through assumptions in November 2015 (see here).

Moreover, previous Bank of England analysis concluded that sterling’s 2008 depreciation induced weaker expenditure switching than the 1992 fall or the 1996 appreciation, with consequently smaller net trade benefits (see chart below).  But rather than blaming weak export markets, the BOE highlighted how UK exporters raised their margins (relative export prices fell by 10pp less than the sterling fall) rather than expanding volumes.  That in turn was driven by service sector exports, which tend to compete on the basis of product quality/reputation rather than price (lower prices can be taken as a signal of poorer quality). And services make up a large share of UK exports (the services trade surplus contrasts with the goods trade deficit).

slide2slide16

Alongside that, the BOE highlighted the constrained ability of UK firms to substitute for some types of imports (now more expensive). Moving into new markets, either domestic or foreign, requires investment.  Unfortunately, UK capex has been weak in the aftermath of the financial crisis and corporate borrowing has been weak.

Moreover, a major channel via which adverse Brexit-related impacts seem likely to occur is via persistent elevated uncertanity causing capex to to delayed or cancelled. There’s already evidence for this in investment surveys and the BOE credit conditions survey reports weak expected corporate credit demand featuring the weakest loan demand to fund capex in five years (see chart above).  Overall, Such an uncertain environment isn’t particularlt conducive to UK firms undertakign the investment to compete in new UK and foreign markets.

Sterling isn’t yet particularly “cheap” on real basis: overshooting may be required

What matters for international trade is the real exchange rate – the nominal rate adjusted for changes in a country’s labour costs/inflation relative to those of its competitors. Expanding export market share (and reducing import penetration) is more likely if the real exchange rate is competitive (“undervalued”).  Unfortunately, although sterling’s nominal depreciation since November is dramatic (£ERI down around 20%), it hasn’t yet caused sterling to be particularly “cheap” on a real exhange rate basis relative to the long-sweep of history.

There are many ways to calculate the “equilibrium” real exchange rate and hence whether the currency has become undervalued. But as a simple start the chart below illustrates that the sterling real effective exchange rate is only a little below it’s long term averages on both a relative unit labour cost (ULC) or relative CPI basis (even when I extend the real exchange rate series for changes in the £ERI to October).

Importantly, neither real £ exchange rate is significantly below their levels in the 1990s or after the 2007/8 financial crisis (only the real ULC based series appears slightly below the post financial crisis lows).  Given the constrained benefits of those previous £ depreciations, there’s little reason to argue that UK net exports are about to surge, especially in a low world trade growth environment in which structural contraints factors have grown (integrated supply chains and specialisation have certainly progressed since 1992, even if they have slowed a little recently).

slide8slide17

Indeed, given these constellation of factors £ may well need to overshoot it’s eventual medium-term level if UK net exports are to provide a major support to GDP growth and hence offset the propsective slowdowns in corporate investment and household consumption (driven by protracted uncertainty and declining real incomes).  In other words, further sterling depreciation seems likely to be necessary.

A small step takes us to FEER-type approaches to determining “equilibrium” exchange rates – which focus (cyclically adjusted) current account deficits. And Brexit puts the UK’s 5.7% of GDP current account deficit, the highest in the G10 (see chart above) into greater focus and suggests that further £ depreciation is required to move the current account deficit to a more-sustainable level. Indeed, Goldman Sachs recently argued that £ remains 10% overvalued on this basis (see here).

Such approaches have the benefit of potentially capturing the important structural break which Brexit represents, unlike many of the filter-based approaches often employed by market participants. Indeed, my previous post argued that a worrying sign was that the current account deficit has recently become increasingly financed by more volatile portfolio inflows (with net FDI inflows drying up). But FX Valuation effects cause me to aim off the Goldman’s headline 10% overvaluation conclusion, via two main mechanisms.

  • First, sterling’s depreciation will raise the sterling value of foreign investment earnings – indeed this seems effect may well provide a more powerful boost to the UK current account than (stickier) net exports. Recall that the UK current account deficit has reached record levels mainly because of weaker net investment income, reflecting lower foreign yields, rather than because of particularly weak UK net exports (see my April post).
  • Second, the currency composition of UK foreign assets and liabilities means that £ depreciations generate improvements in the UK’s international investment position (i.e.net foreign assets).  Such valuation effects This helps explain why the UK IIP has not deteriorated sharply despite persistsnt

Unfortunately, we don’t get sight of the Q3 current account data until 22 December. But it’s possible that sterling’s depreciation may raise the attractivenes of some UK assets to foreign suitors e.g. following Softbank’s acquisition of ARM.

Near-term £ stability, potential sell rallies  strategies

Sterling’s relative stabilisisation in recent days has been underpinned by traders’ hopes that Government concessions to Parliament mean that hard Brexit isn’t inevitable. Plus Chancellor Hammond has rowed back on PM May’s ill-advised apparent attack on BoE independence (BoE Governor Carney today graciously extended the olive branch).  But May’s concessions to Parliament aren’t substantive: it will be hard for parliament to reject a proposed Brexit deal close to it’s signing, given the wish to avoid the provoking further uncertainty and market volatility.

slide1slide18

But it’s notable that the continued £ weakness has coincided with 10 year gilt yields spiking sharply (see chart).  This validating my earlier views that risk risk premia on £ assets were set to rise (see my April post as well as earlier this month here). The spike in gilt yields has, importantly, lagged the rise in UK inflation breakevens (which could be starting to concern the MPC) thereby leaving the UK with the lowest real interest rates in the G10 (see Chart) and hence undermining any potential support to the £ generated by higher yields.  The bottom line is that yields are rising for “bad” reasons and £ is most vulnerable to rising risk premia in the near term.  And, as Governor Carney noted, such risk premia effects can only rise if politicians keep taking pot shots at BoE independence (hopefully the Government will internaluse this lesson).  Moroever, £ implied volatilities remain elevated, even if substantially below their levels around the referendum.

So overall I stick to my previous view that sterling will likely fall further – especially once reality bites on the difficulty confronting Article 50 negotiations and Brexit’s slow-burn adverse economic impacts become more apparent in the data (weak UK corporate investment and hiring, adverse hit to government finances).

Conversely, a November MPC rate cut now looks less likely than it did a few weeks ago, if not impossible, given that near-term data haven’t fallen off a cliff (Thursday’s preliminary Q3 GDP release is likely to be OK given relatively-robust services sector output) and inflationary pressures are mounting (which would be positive after several years of undershooting the target were it not for the likely adverse impact on household real incomes). And squeezes of extended short £ positions can’t be ruled out (although I noted here that they had been resilient to mini risk-off periods).

So in the near-term investors may well shift to selling £ rallies and £ could well trade sideways. That said, £/$ could fall below 1.20 on a December FOMC rate rise (around 70% priced) but €/£ could fall back should the 4 Dcember Italian referendum produce a “no” vote (the polls are too close to call) or should the ECB extend QE in December. Draghi last week offered investors hope of further action in December when the phrase “there are no signs yet of a convincing upward trend in underlying inflation.” was added to the ECB opening statement and Draghi noted that the anticipated rise in inflation is “predicated on current financing conditions”.  But Draghi was (unsurprisingly) unforthcoming on the work of the Committees examinng how to ensure continued smooth implementation of ECB measures and recent better EA data (PMIs and German IFO survey  this week) mean that a QE extension isn’t a done deal.

Advertisements

3 thoughts on “UK trade to party like it’s 1992? Watch out cyclical and structural headwinds

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s