Global Investment Outlook 2015 — Divergence

The years following the 2008 financial crisis had a simple organising principle: major central banks were all easing. Rates were low or zero, balance sheets were expanding, and the direction of travel was the same nearly everywhere. Investors could hold a single view about policy and apply it globally.

2015 ended that. The US Federal Reserve moved toward its first tightening since the crisis, while the European Central Bank and the Bank of Japan expanded stimulus. For the first time in roughly seven years, the world's major monetary authorities were moving in opposite directions.

Policy divergence between countries expresses itself through exchange rates. Capital moves toward the currency offering higher returns, strengthening it and weakening the others. That single mechanism drove most of what happened in 2015 and much of what followed.

Two other events compounded it. Oil fell dramatically, in a decline driven more by supply than by demand — a distinction that mattered enormously for interpreting it. And China adjusted its currency in August in a move that was modest in size but significant in what it revealed about the authorities' priorities.

Underneath these macro events, something quieter was happening in private markets. Valuations for late-stage private technology companies had risen to levels that raised, for the first time in that cycle, a serious question about whether private marks reflected what public markets would pay.

Why divergence transmits through currency

The mechanism deserves stating plainly because it explains most of 2015's cross-asset behaviour.

Capital seeks the highest risk-adjusted return. If one country's central bank raises rates while another's cuts, assets denominated in the first currency become relatively more attractive. Capital flows toward it. That flow is itself a purchase of the currency, which strengthens it.

A strengthening dollar produces a chain of consequences that reach well beyond currency markets:

• Dollar-denominated debt owed by non-US borrowers becomes harder to service, because the borrower earns in a currency that has weakened against the debt. This is the primary channel through which US monetary policy transmits to emerging markets.
• Commodities priced in dollars face downward pressure, since a stronger dollar means the same commodity costs more in every other currency, dampening demand.
• US exporters lose competitiveness while importers gain.
• Foreign earnings of US multinationals translate into fewer dollars, reducing reported profits without any change in underlying business.

The emerging-market channel was the most consequential. Many companies and governments outside the US had borrowed in dollars during the low-rate years, when doing so was cheap and the dollar was weak. When the dollar strengthened, the real burden of that debt rose without any new borrowing.

A country can experience a tightening of financial conditions without its own central bank doing anything, simply because it borrowed in someone else's currency.

Oil: a supply story misread

Oil fell sharply through 2015. The interpretation of that decline mattered more than the decline itself, and much commentary got it wrong.

A fall in oil prices can have two very different causes. Weak demand signals a slowing global economy and is genuinely bad news, because oil demand tracks industrial activity. Abundant supply means more oil is being produced, which is a transfer of wealth from producers to consumers rather than a signal about growth.

2015's decline was substantially a supply story. US shale production had expanded significantly over preceding years, and OPEC declined to cut output to defend price — a strategic decision to maintain market share rather than support prices, which meant allowing the oversupply to persist.

The distinction determined which consequences were durable:

• Energy producers and oil-exporting economies suffered directly and lastingly.
• Energy-importing economies and consumers received what amounted to a tax cut.
• Credit markets came under stress specifically in energy-related high yield, which had grown substantially to finance shale development — a sector-specific problem widely misread as a broad credit signal.
• Inflation fell, complicating central bank decisions in economies where inflation was already below target.

Reading the decline as a demand signal implied a global slowdown. Reading it as supply implied a redistribution. The second reading proved closer to correct, and investors who acted on the first drew the wrong conclusions about growth.

China's August adjustment

In August 2015 Chinese authorities adjusted the mechanism by which the currency's daily reference rate was set, resulting in a depreciation against the dollar.

The magnitude was small. The reaction was not. Understanding why requires distinguishing what happened from what it signalled.

Markets had operated on an assumption that Chinese authorities would maintain currency stability, treating it as a fixed feature of the landscape. The adjustment did not disprove that entirely, but it demonstrated the assumption was a policy choice rather than a constant — and policy choices can change.

That prompted a reassessment across several dimensions at once. Whether the Chinese economy was weaker than official figures suggested, if authorities felt the need to support exports. Whether further depreciation would follow, prompting capital flight. And whether other exporting economies would respond in kind.

The episode illustrates something general about how markets process information: the significance of an event is not proportional to its magnitude but to how much it changes the distribution of expected outcomes. A small move that invalidates an assumption can matter more than a large move that confirms one.

It also connected to the divergence theme. A currency managed against the dollar imports US monetary policy. As the Fed moved toward tightening and the dollar strengthened, maintaining that link meant importing a tightening China's own conditions did not warrant. The adjustment was, in part, a response to that pressure.

Private valuations begin to detach

Away from the macro events, 2015 saw the first serious discussion of a problem that would take years to fully surface.

Late-stage private technology valuations had risen substantially, and a cohort of private companies valued above a billion dollars had emerged in sufficient numbers to acquire a label. The question raised — sceptically at the time, correctly in hindsight — was whether those valuations reflected what public markets would actually pay.

Several structural features made detachment possible:

• Private rounds are negotiated, not auctioned. A price is set between a company and a small number of investors, without the continuous price discovery a public market provides.
• Structure can inflate headline valuations. Terms like liquidation preferences and ratchets protect the investor's downside, which makes a higher nominal price rational for them while making that price a poor measure of what the whole company is worth.
• There was no forcing mechanism. A company could remain private indefinitely, and capital availability meant it never had to test its valuation against a public bid.

The gap that opened in this period was not resolved quickly. It persisted, widened through subsequent years, and was still being worked through when the 2019 listings of several large private companies finally forced the comparison — a theme the 2019 report picks up directly.

How 2015 was misread at the time

Contemporaneous interpretation got several things wrong in ways that were expensive, and the errors are instructive because they recur.

The oil decline was read as a growth signal. This was the largest error and it followed from a reasonable heuristic: oil demand tracks industrial activity, so falling oil usually means slowing growth. The heuristic failed because the causation ran the other way — supply had expanded. Investors who inferred a global slowdown positioned defensively into a period when energy-importing economies were receiving a substantial terms-of-trade benefit.

China's currency adjustment was read as a devaluation strategy. Much commentary treated the August move as the opening of a deliberate campaign to depreciate for export advantage. The subsequent years did not bear that out — the authorities spent substantial reserves resisting depreciation, which is the opposite of what a devaluation strategy implies. The move is better read as a response to the pressure of maintaining a dollar link while the dollar strengthened.

Divergence was read as temporary. A widespread view held that other major central banks would follow the Fed within a year or two, restoring the synchronised policy environment. They did not, and the divergence persisted in various forms for most of the following decade.

The private valuation question was read as a bubble question. The framing at the time was binary: were unicorn valuations a bubble that would burst? That framing obscured the more useful observation the 2015 US venture report develops — that private and public valuations are produced by different processes and measure different things. A gap between two different measurements is not the same as a bubble, and it does not resolve by bursting. It resolves when a forcing mechanism arrives, which took four more years.

The common thread in the first three is a preference for interpretations that preserve the existing framework. A supply-driven oil decline, a persistent policy divergence, and a currency adjustment forced by circumstance are all harder to hold than their alternatives, because each requires abandoning a mental model that had been working.

What an allocator could act on

The mechanisms in this report translate into observable positions rather than forecasts, which is the useful test of whether an analysis is actionable.

The dollar-debt exposure was measurable in advance. The BIS publishes dollar-denominated debt by borrower country, free and quarterly. An investor holding emerging market exposure in 2015 could have ranked their holdings by this measure and understood which faced imported tightening. This was not a forecast — it was a lookup.

The oil interpretation was testable. The supply-versus-demand question can be assessed from production data rather than from price. US production volumes and OPEC output decisions were published. An investor who checked whether supply had risen before concluding demand had fallen would have reached the correct interpretation from public data.

Currency hedging policy was the decision that mattered most, and it was frequently not treated as a decision at all. An unhedged foreign holding in 2015 experienced the currency move directly, and for several markets the currency effect exceeded the equity effect. The most consequential choice available to a foreign investor in the region that year was whether to hedge, and it is a choice that is often made by default rather than deliberately.

The private valuation gap suggested a diligence question rather than a market call. The actionable response was not to avoid late-stage private investment but to ask what the headline valuation actually represented — specifically, what preference terms sat underneath it and what the implied common-share value was. That question is answerable from the term sheet and was frequently not asked.

The pattern across all four: the useful response to 2015 was a set of questions with checkable answers, not a view about direction. That is generally true of structural shifts, and it is why they reward analysis over conviction.

What 2015 established

• Policy divergence became a permanent analytical category after seven years in which synchronised easing made it unnecessary.
• Dollar strength was re-established as a global financial condition, not merely a currency-market phenomenon.
• The supply-versus-demand distinction in commodities was demonstrated to matter more than the price move itself.
• Assumptions about policy stability were shown to be assumptions, most visibly regarding China's currency.
• The private-public valuation gap opened, beginning a structural issue that shaped the following decade.

Why divergence persisted longer than expected

A footnote worth recording, because it shaped the following decade more than the initial divergence did.

The consensus expectation in 2015 was that divergence would be temporary — that other major central banks would follow within a year or two and the synchronised environment would resume. That expectation was reasonable and it was wrong for reasons that became clearer later.

Underlying conditions differed more than the policy stance suggested. The US had completed more of its post-crisis balance sheet repair, had more favourable demographics, and had a banking system that had recapitalised faster. Europe and Japan were addressing structurally lower growth and structurally lower inflation, which are not conditions that resolve on a policy timetable.

Inflation dynamics differed. The US approached its target; Europe and Japan did not, and had not for an extended period. A central bank persistently undershooting cannot tighten regardless of what others do.

The consequences of persistence:

• The dollar's strength was sustained rather than temporary, which meant the emerging market debt burden described above compounded rather than reversed.
• Yield differentials became a structural feature, driving persistent cross-border capital flows and supporting the currency hedging costs that became a material consideration for institutional portfolios.
• The institutional yield problem in Europe and Japan intensified, which is the mechanism the 2016 report describes pushing capital toward private markets.

The general lesson concerns the default assumption of convergence. Markets reliably expect divergent conditions to converge, because convergence is the historical norm over long periods and because a persistent divergence is harder to model. But convergence is a claim about underlying conditions, not about policy — and where the underlying conditions differ structurally, the policy divergence persists as long as they do.