The end of the dollar’s exorbitant privilege

A crash is likely given the collapse in US domestic saving and a gaping current account deficit

Stephen Roach 


The riddle once posed in the 1960s by former French finance minister (eventually president) Valéry Giscard d’Estaing is about to be solved. Giscard bemoaned a US that took advantage of its privileged position as the world’s dominant reserve currency and drew freely on the rest of the world to support its over-extended standard of living. 

That privilege is about to be withdrawn. A crash in the dollar is likely and it could fall by as much as 35 per cent by the end of 2021.

The reason: a lethal interplay between a collapse in domestic saving and a gaping current account deficit. In the second quarter of 2020, net domestic saving — depreciation-adjusted saving of households, businesses and the government sector — plunged back into negative territory for the first time since the global financial crisis. 

At -1.2 per cent in the second quarter, net domestic saving as a share of national income was fully 4.1 percentage points below the first quarter, the steepest quarterly plunge in records that go back to 1947. 

 Unsurprisingly, the current account deficit followed suit. Lacking in saving and wanting to grow, the US levered its exorbitant privilege to borrow surplus saving from abroad. 

That pushed the current account deficit to -3.5 per cent of gross domestic product in the second quarter — 1.4 percentage points below that in the first period and also the sharpest quarterly erosion on record.

While a Covid-related explosion in the federal government deficit is the immediate source of the problem, this was an accident waiting to happen. 

Going into the pandemic, the net domestic saving rate averaged just 2.9 per cent of gross national income from 2011 to 2019, less than half the 7 per cent average from 1960 to 2005. 

This thin cushion left the US vulnerable to any shock, let alone Covid.

As budget deficits pile up in the years ahead, further downward pressure on domestic saving and the current account will intensify. 

The latest estimates of the Congressional Budget Office put the federal deficit at 16 per cent of gross domestic product in 2020 before receding to “just” 8.6 per cent in 2021. 

Assuming the US Congress eventually agrees to another round of fiscal relief, a much larger deficit for 2021 is likely.

This will take the US net saving rate far deeper into negative territory than during the global crisis That has ominous implications for America’s future. 

After setting aside depreciation required of an ageing capital stock of buildings and infrastructure, the US is, in effect, liquidating the net saving required for the expansion of productive capacity. 

Without borrowing surplus saving from abroad, growth becomes impossible. 

The current account deficit will only deepen as a result.

That’s when the dollar loses its special privilege. With America’s position as the world’s dominant reserve currency slowly eroding since 2000, foreign lenders are likely to demand concessions on the terms for such massive external financing. This normally takes two forms — an interest rate and/or a currency adjustment. 

The Federal Reserve has recently shifted to a strategy that takes into account an average of inflation rather than a specific target, and promised to keep policy rates near zero for several more years. 

That means the interest rate channel has effectively been closed. As a result, more of the current account adjustment will now be forced through a weaker dollar.

The US dollar’s lofty value makes it especially vulnerable. Despite recent falls, a broad index of the dollar’s real effective exchange rate remains some 27 per cent above its July 2011 low. 

That leaves the greenback as the world’s most overvalued major currency, just as the US gets sucked into an unprecedented savings-current account vortex.

Currencies are relative prices. The dollar has always benefited from the seductive charm of TINA — that there is no alternative. Think again. 

The July 21 agreement on a Next Generation EU Fund of €750bn ($858bn) finally establishes a pan-European fiscal policy. That should boost the undervalued euro. The renminbi, gold and cryptocurrencies are also alternatives to the once invincible dollar.

The dollar index fell 33 per cent in real terms both in the 1970s and the mid-1980s, and another 28 per cent from 2002 to 2011. 

During those three periods, the net domestic saving rate averaged 4.9 per cent (versus -1.2 per cent today) and the current account deficit was -2.5 per cent of gross domestic product (versus -3.5 per cent today). 

With the US having squandered its exorbitant privilege, the dollar is now far more vulnerable to a sharp correction. A crash is looming.

Why Biden Is Better Than Trump for the Economy

The presumption that Republicans are better than Democrats at economic stewardship is a longstanding myth that must be debunked. For all Americans who care about their and their children’s future, the right choice this November could not be clearer.

Nouriel Roubini

NEW YORK – Joe Biden has consistently held a wide polling lead over US President Donald Trump ahead of November’s election. But, despite Trump’s botched response to the COVID-19 pandemic – a failure that has left the economy far weaker than it otherwise would have been – he has maintained a marginal edge on the question of which candidate would be better for the US economy. 

Thanks to Trump, a country with just 4% of the world’s population now accounts for more than 20% of total COVID-19 deaths – an utterly shameful outcome, given America’s advanced (albeit expensive) health-care system.

The presumption that Republicans are better than Democrats at economic stewardship is a longstanding myth that must be debunked. In our 1997 book, Political Cycles and the Macroeconomy, the late (and great) Alberto Alesina and I showed that Democratic administrations tend to preside over faster growth, lower unemployment, and stronger stock markets than Republican presidents do.

In fact, US recessions almost always occur under Republican administrations – a pattern that has persisted since our book appeared. The recessions of 1970, 1980-82, 1990, 2001, 2008-09, and, now, 2020 all occurred when a Republican was in the White House (with the exception of the double-dip recession of 1980-82, which started under Jimmy Carter but continued under Ronald Reagan). Likewise, the Great Recession of 2008-09 was triggered by the 2007-08 financial crisis, which also occurred on the GOP’s watch.1

This tendency is not random: loose regulatory policies lead to financial crises and recessions. And, compounding matters, Republicans consistently pursue reckless fiscal policies, spending as much as Democrats do, but refusing to raise taxes to make up for the resulting budget shortfalls.1

Owing to such mismanagement under the George W. Bush presidency, President Barack Obama and Vice President Biden inherited the worst recession since the Great Depression. In early 2009, the US unemployment rate surpassed 10%, growth was in free fall, the budget deficit had already exceeded $1.2 trillion, and the stock market was down almost 60%. Yet, by the end of Obama’s second term in early 2017, all of those indicators had massively improved.1

In fact, even before the COVID-19 recession, US employment and GDP growth, as well as the stock market’s performance, were better under Obama than under Trump. Just as Trump inherited millions from his father, only to squander it on business failures, so he inherited a strong economy from his predecessor, only to wreck it within a single term.

The rally in equity prices this past August coincided with a hardening of Biden’s polling lead, suggesting that markets are not nervous about a Biden presidency, or about the prospects of a Democratic sweep of Congress. The reason is simple: a Biden administration would be unlikely to pursue radical economic policies. 

Biden may be surrounded by progressive advisers, but they are all fully within the political mainstream. Moreover, his vice-presidential pick, US Senator Kamala Harris of California, is a proven moderate, and most of the Democratic senators who would be seated in a new Congress are more centrist than the left wing of their party.

Yes, a Biden administration might raise marginal tax rates on corporations and the top 1% of households, which Trump and congressional Republicans cut merely to give wealthy donors and corporations a $1.5 trillion handout. But a higher tax rate would result in only a modest hit to corporate profits. 

And any costs to the economy would be more than offset by closing the loopholes that allow for tax avoidance and shifting profits and production abroad, and with Biden’s proposed “Made in America” policies to bring more jobs, profits, and production home.1

Moreover, while Trump and his fellow Republicans have not even bothered to formulate a policy platform for this election, Biden has proposed a suite of fiscal policies designed to boost economic growth. If Democrats take control of both houses of Congress and the White House, a Biden administration would pursue a larger fiscal stimulus targeted at households, workers, and small businesses that need it, as well as job-creating infrastructure spending and investments in the green economy. 

They would not invest in tax cuts for billionaires, but rather in education and worker retraining, and in proactive industrial and innovation policies to ensure future competitiveness. Private business would no longer be terrorized by the president in Twitter tantrums.

Democrats also are calling for higher minimum wages to boost labor income and consumption, along with more sensible regulations to reduce carbon dioxide emissions. They would push for policies to restore some bargaining power to workers, and to protect savers from predatory financial institutions. 

And they would have a much more sensible approach to trade, immigration, and foreign policy, repairing US alliances and partnerships and pursuing a policy of “coopetition” rather than lose-lose containment vis-à-vis China. All these measures would be good for jobs, growth, and markets.

Although Trump ran as a populist, he is a wannabe plutocrat – a pluto-populist – and that is how he has governed. His economic policies have been disastrous for US workers and long-term economic competitiveness. Trade and immigration policies that were billed as measures to restore US jobs have had the opposite effect. 

The “deaths of despair” that disproportionately afflict white blue-collar and precariat workers have not fallen under Trump; with more than 70,000 drug overdose deaths in 2019, this American carnage continues. If the US is to fill the high-value jobs of the future, it will need to train its labor force, not embrace self-destructive protectionism and xenophobia.

The choice for US voters who are concerned about America’s economic prospects could not be clearer. Biden, who has long tapped into blue-collar concerns, is the only presidential candidate in recent history without an Ivy League background. 

He has a better chance than anyone of rebuilding the Democratic coalition and winning back the support of disaffected, working-class voters. For all Americans who care about their and their children’s future, the right choice this November could not be clearer.

Nouriel Roubini, Professor of Economics at New York University's Stern School of Business and Chairman of Roubini Macro Associates, was Senior Economist for International Affairs in the White House’s Council of Economic Advisers during the Clinton Administration. He has worked for the International Monetary Fund, the US Federal Reserve, and the World Bank. His website is, and he is the host of

Why Low Interest Rates Hurt Retirees

Retirees took another wallop from COVID-19 with the Federal Reserve’s announcement two weeks ago that it expects to hold interest rates near zero at least until 2023 because of the pandemic. 

That spells lower returns for retirement accounts, and it adds to the underfunding of pensions that has worried retirees for many years now.

The implications of lower returns on investments are that retirees may save less, dip into their retirement savings, and start collecting Social Security benefits earlier than planned, according to Olivia S. Mitchell, Wharton professor of business economics and public policy and executive director of the School’s Pension Research Council.

“Low returns from the market are essentially a tax on retirees,” Mitchell said in a recent episode of the Wharton Business Daily show on SiriusXM. (Listen to the podcast above.)

“In the good old days, people used to ladder their bonds, put a little bit of money in the market, and try to live off those returns,” Mitchell noted. (A bond ladder refers to investments in bonds with varying maturities so that a portfolio does not get locked into one type of bond.) 

“This is not feasible any longer. In fact, it’s even worse, because those lower nominal returns are in many cases negative in real, or inflation-corrected, returns.”

Retirees may respond to the prospect of low returns by saving less, Mitchell said: “If you’re not rewarded for deferring your consumption as much, then why do it?” Their savings would fall especially in tax-qualified retirement plans, she predicted. The tax-qualified feature is helpful for those who aim to build an asset base over time with interest and other returns on their investments, while they are in relatively lower tax brackets. 

But now, “those build-ups are simply not happening the way that people had planned,” she said. “If people do save, they’ll probably save less overall, and they will tend to save in other non-tax favored accounts like bank saving and checking accounts, where you’re lucky if you’re earning half a percentage point.”

Desperate Times

Mitchell recalled that the 2008–2009 global financial crisis was “a bath of cold water for retirees, savers, pension funds, insurance companies, and so on.” However, back then, an economic recovery followed, and “the labor market didn’t suffer as badly as it has during the COVID-19 pandemic, and for as long as it has,” she noted. 

Nowadays, people’s perspectives on retirement have changed and they are looking to work a little longer. “But the question is, can you even find a job, especially if you are an older worker?” she asked.

In times of desperation, some people may claim their Social Security benefits sooner than they may have planned — despite the fact that every year of delay could boost their eventual benefit by 7% to 8%, Mitchell said. “People who don’t have any retirement savings may have to go ahead and claim their [Social Security] benefits early, thereby experiencing a lower payout the rest of their lives.”

Others that have lost jobs have drawn down some of their 401(k) savings. Since April, they have been living off the government’s “economic impact payment” of $1,200 per individual (an additional $500 for each child), and the expanded unemployment insurance benefits contained in the $2.2 trillion CARES Act. 

“Those have now tapered off, and Congress has not yet been able to come in with a new COVID financing bill,” Mitchell stated. Some people might consider options like starting their own small business, but “this is a pretty tough environment in which to start a new business,” she added.

The 2020 CARES Act permitted individuals to make early withdrawals up to $100,000 from 401(k) and 403(b) accounts without penalties. That hasn’t caught on so far because the economic stimulus payments provided money to cope with the pandemic in the short term. 

However, it may not be possible to stave off early withdrawals from retirement accounts for too long, said Mitchell.

“Going into the fall and winter, I do worry that [early withdrawals from retirement accounts] will become more of an option if the labor market doesn’t recover,” Mitchell said. “And so, people might end up biting off their nose to spite their face. Yes, they’ll get some cash, but what does it say about their retirement? Not much [that is] good.”

“The first and most important thing that needs to be done by policymakers is to bring Social Security back into solvency.”

A Perfect Storm

Also looming is the possibility that the Social Security Trust Fund could run out of money by 2029, rather than 2032 or 2034 as had been predicted by the Penn Wharton Budget Model after the pandemic struck. This could happen “since people aren’t paying the payroll taxes needed to keep the system afloat and potentially because people are claiming [their benefits] earlier,” said Mitchell. 

“The first and most important thing that needs to be done by policymakers is bring Social Security back into solvency.”

For sure, pension funds and insurance companies have also been suffering from low returns, Mitchell continued. 

“Many of the state and municipal pension plans are probably not going to make it through this COVID crisis with any healthy amount of funding.”

As it happens, pension plans are facing “a perfect storm” now, said Mitchell. 

They faced the 2008–2009 financial crisis without being fully funded, but in later years “continued to invest in risky assets, hoping to make it up in the great capital market lottery.” 

During the pandemic, many pension funds lost 30% to 40% of their value, and the forecasted low returns will make it “very difficult for them to survive,” she added.

In that seemingly hopeless situation, Mitchell saw annuities as a “potentially appealing” option for retirement planning. Annuities are insurance products that pay an income in retirement. 

Even if the insurance investments held by annuity providers don’t make much money, investors who outlive others in their pool will be eligible for survival credits (also known as mortality credits), she noted. 

She suggested that it is a good idea for employers and retirement plan sponsors to include annuities in 401(k) and 403(b) accounts, now permitted by the Secure Act since late 2019.

Forging a Stronger Post-Pandemic ASEAN+3 Economy

The unprecedented challenge that COVID-19 poses to the ASEAN+3 countries further underscores the importance of regional financial cooperation. To that end, recent enhancements to the region's currency-swap arrangement will help to mitigate Asian economies' vulnerability to economic and financial shocks.

Aso Taro, Le Minh Hung

TOKYO/HANOI – The modern international financial system emerged from the devastation of World War II. Since then, it has continued to be shaped by historic slumps – most recently, the 2008 global financial crisis.

Today, the COVID-19 pandemic is putting the global financial system to another stringent test. And the unprecedented challenge facing the ASEAN+3 region – the ten members of the Association of Southeast Asian Nations (Brunei Darussalam, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, the Philippines, Singapore, Thailand, and Vietnam), plus China, Japan, and Korea – further underscores the importance of regional financial cooperation.

Since the 1997 Asian financial crisis, the ASEAN+3 economies have been diligently enhancing their regional financial-safety measures. The annual ASEAN+3 Finance Ministers’ and Central Bank Governors’ Meeting has been the focal point for regional cooperation aimed at strengthening economic and financial resilience.

To this end, the ASEAN+3 countries decided in 2000 to establish the Chiang Mai Initiative, the region’s financial safety net. The CMI was a network of bilateral swap arrangements among the ASEAN+3 countries that aimed to provide US dollar liquidity to members in times of need and supplement financial assistance from the International Monetary Fund.

In March 2010, following the global financial crisis, the CMI evolved into the Chiang Mai Initiative Multilateralization (CMIM) Agreement, under which the participating swap arrangements would henceforth be governed by a single agreement and a centralized decision-making body. The size of the CMIM facility was set at $120 billion.

Since its establishment, the CMIM Agreement has been upgraded twice. The first amendment, in 2014, doubled the size of the arrangement to $240 billion, and increased the “IMF de-linked” portion – the maximum amount members could access without IMF co-financing – from 20% to 30%.

The second amendment, to promote greater flexibility in co-financing with IMF financial assistance, and strengthen coordination with the Fund, entered into force in June.

At their most recent meeting, on September 18, ASEAN+3 finance ministers and central-bank governors exchanged views on the global and regional economic outlook, as well as policy responses to risks and challenges arising from the COVID-19 pandemic. 

Over the past few months, the region’s policymakers have deployed extraordinary pandemic-related measures in the form of targeted fiscal, monetary, and credit support to households and firms, and have afforded regulatory forbearance and liquidity support to the financial system.

Against this backdrop, ASEAN+3 finance ministers and central-bank governors announced a timely and historic agreement to strengthen the CMIM, which will certainly help members to cope better with the heightened risk and uncertainty posed by the pandemic. 

The latest enhancement will allow members to access up to 40% of the CMIM facility without IMF co-financing. Moreover, members agreed to formulate the option of using their own currencies for CMIM crisis financing, in addition to the dollar, on a voluntary and demand-driven basis.

These amendments, which will take effect upon completion of the signing process by all ASEAN+3 members, further strengthen the CMIM as a robust and reliable regional self-help mechanism, and reinforce its importance and relevance in the global financial safety net.

In this regard, the ASEAN+3 Macroeconomic Research Office (AMRO), established in 2011 to conduct regional macroeconomic surveillance and aid the implementation of the CMIM, plays an important role as a “trusted family doctor” to support the group’s members. 

AMRO’s mandate to contribute to regional economic and financial stability is even more critical in today’s environment. 

Under the leadership of Doi Toshinori, AMRO has produced timely analyses and updates on the pandemic’s regional impact to support members’ policymaking.

Since its outbreak, the COVID-19 pandemic has taken a heavy toll on both human life and national economies. Within the ASEAN+3 region, policymakers have implemented necessary containment measures to control the transmission of the coronavirus, which inevitably affected economic activity. 

With the global economy increasingly interconnected, the disruptions to international supply chains and the challenges faced by many industries have demonstrated the importance of mitigating the region’s vulnerability to economic and financial shocks. 

We are encouraged by the progress of the CMIM and AMRO as important tools for strengthening our economies’ resilience.

While growth is projected to fall sharply for many economies this year, we expect the ASEAN+3 economies to rebound in due course – and there are early signs of recovery. 

Given the pandemic’s uncertain trajectory, however, the ASEAN+3 countries will remain vigilant, and we will cautiously plan our exit from pandemic-related measures to safeguard growth and financial stability in the region.

Finally, the ASEAN+3 finance ministers and central-bank governors pledge to remain resolute in our commitment to uphold an open and rules-based multilateral trade and investment system, and to strengthen regional cooperation and integration.

Aso Taro, a former prime minister of Japan (2008-09), is Japan’s Deputy Prime Minister and Minister of Finance.

Le Minh Hung is Governor of the State Bank of Viet Nam.

Match and mix

How hybrids have upturned evolutionary theory

The origin of species is more complex than Darwin envisaged

In 1981 peter and Rosemary Grant, a husband-and-wife team of evolutionary biologists, spotted something odd on Daphne Major. Every year for the previous decade they had travelled from Princeton University to this island in the Galápagos, to study its three endemic tanager species, part of a group known colloquially as “Darwin’s finches”. 

On this occasion their eyes were drawn to an unusual male that sported dark feathers and sang a unique song. Genetic analysis later identified him as a large cactus finch, probably blown in from Española, another part of the archipelago that is over 100km away.

Intrigued, the Grants followed the castaway as he explored his new home. They watched him mate with a local female medium ground finch. That produced five fit, healthy offspring. Those offspring were also surprisingly sexually selective. A single male excepted, they and their descendants mated only among themselves—and they have continued to do so ever since.

Despite this heavy inbreeding, the hybrids (two of which are pictured above) have been successful. They have carved out a niche in which they use their size and their deep beaks to exploit the large woody fruits of the Jamaican feverplant, which grows locally. 

They have, to all intents and purposes, become another species of Darwin finch, of which 13 were previously recognised. Though they do not yet have a Latinised scientific name, they are known to all as the “Big Bird” lineage.

Heretical thinking

This story would once have been considered deeply implausible. Evolution’s orthodox narrative does not suggest that hybridisation is how new animal species emerge. But, as genetic testing has proliferated, biologists have been confronted with an unexpected fact. Hybrids are not an evolutionary bug. They are a feature.

That knowledge is changing the way people think about evolution. The neat family trees envisaged by Charles Darwin in one of his early notebooks (see picture below) are turning into webs, and the primacy of mutation in generating the variation which natural selection then winnows is being challenged. 

The influx of genes accompanying hybridisation creates such variation too—and the harder people look, the more important that seems to get. Hybridisation also offers shortcuts on the long march to speciation that do not depend on natural selection at all. As the example of the Big Bird lineage shows, instead of taking millennia to emerge, a new species can appear almost overnight.

        As Darwin saw it

In truth, all this had already been recognised for simple organisms like bacteria. These exchange genes promiscuously between both more and less related individuals. But bacteria were unknown when Darwin came up with natural selection, and, ever since then, the subject of speciation has been dominated by examples drawn from animals and plants. 

To recognise that what is true for bacteria is also true for these multicellular organisms has profound implications, not least for how human beings understand their own origins. It seems appropriate, then, that the birds whose diversity helped inspire Darwin still have evolutionary tales to tell.

The conventional view of evolution is that mutations happen at random. Maladaptive ones are then eliminated by competitive pressure while adaptive ones proliferate. The result, over long periods of time and assisted by populations sometimes being split up by external circumstances, is change which eventually crystallises into new and separate species.

That process does leave the door open to hybrids. The genomes of closely related species may remain sufficiently similar to produce viable offspring. But these genes often fit together less well than those of parents from the same species. As a consequence, even viable hybrids are frequently infertile (think mules) and are also at higher risk of developmental and other types of illnesses. 

In fact, infertility in male hybrids is so common that it has a name—Haldane’s rule. 

This sort of thing was enough to persuade most of Darwin’s 20th-century disciples that the need to avoid hybridisation was actually a driving force which caused natural selection to erect reproductive barriers between incipient species, and thus encouraged speciation.

There is, though, another way of looking at hybridisation. Mixing the traits of two parent species might actually leave their hybrid offspring better off. This is called hybrid vigour, or heterosis. 

The interplay of two species’ genes can even produce traits displayed by neither parent. 

This is known as transgressive segregation and the resulting hybrid may be surprisingly well adapted to a completely new niche, as was the case with the Big Birds.

Both the maleficent and beneficent effects of hybridisation are real. The question is, which wins out more often in practice? 

In plants, it is frequently the beneficent. This is a consequence of plants’ unusually malleable genetics. 

The nuclear genomes of complex organisms (animals, plants, fungi and single-celled organisms such as amoebae) are divided into bundles of dna called chromosomes. Such organisms are generally either haploid or diploid, meaning that each cell nucleus contains either one or two copies of every chromosome. 

Human beings are diploid. They have 23 chromosomal pairs, for a total of 46 individual chromosomes. But there are exceptions. 

Plants, for instance, are frequently polyploid—meaning that each nucleus contains copies in greater multiples than two. 

To take one example, Californian coastal redwoods have six copies. Since redwood cell nuclei have 11 distinct types of chromosome, they host a total of 66 chromosomes altogether.

Sometimes, polyploidy is a result of an organism’s genome spontaneously doubling. 

Often, though, it is a consequence of hybridisation, with the chromosomes of both parents ending up in a single nucleus. However it arises, polyploidy provides spare copies of genes for natural selection to work on while other versions of them continue with their original function. 

And if it is also the result of hybridisation, it brings the additional possibilities of heterosis and transgressive segregation.

On top of this, by changing an organism’s chromosome count polyploidy has another pertinent effect. It creates an instant barrier to breeding with either parent species. 

That gives a new, incipient species a chance to establish itself without being reabsorbed into one of the parental populations. The results can be spectacular. Recent evidence suggests, for example, that hybridisation between two plant species in the distant past, followed by a simple doubling of the number of chromosomes in their offspring, may be responsible for much of the extraordinary diversity in flowering plants that is seen today.

Plants seem to be easy beneficiaries of hybridisation. For many animals, however—and for mammals in particular—extra chromosomes serve not to enhance things, but to disrupt them. Why, is not completely clear. Cell division in animals seems more easily confounded by superfluous chromosomes than it is in plants, so this may be a factor. 

Plants also have simpler cells, which are more able to accommodate extra chromosomes. Whatever the details, animal hybrids appear to feel the effects of genetic incompatibility far more acutely than do plants, and are therefore less able to benefit from heterosis. 

Evolutionary biologists therefore assumed for a long time that hybridisation played a negligible role in animal evolution—and there was little evidence to suggest otherwise.

Advances in dna sequencing have changed that by letting people look under the bonnet of evolutionary history. This has uncovered a steady trickle of animals breathed into life entirely by hybrid speciation. 

They include some familiar names. The European bison, for instance, is the result of hybridisation, over 120,000 years ago, between two now extinct species—the ice-age steppe bison and the auroch. The latter were the wild antecedents of modern domestic cattle, and survived in Jaktorow Forest, in Poland, until 1627.

Something similar is true of the Atlantic Clymene dolphin. Genetic analysis has revealed that this cetacean, which roams the briny between west Africa, Brazil and the Gulf of Mexico, owes its existence to a hybridisation that happened between two globe-trotting others, the striped dolphin and the spinner dolphin.

At least one hybrid animal, moreover, traces its ancestry to three species. Genetic analysis shows that Artibeus schwartzi, a Caribbean fruit bat, is a result of hybridisation, within the past 30,000 years, of the Jamaican fruit bat (Artibeus jamaicensis), the South American flat-faced fruit-eating bat (Artibeus planirostris) and a third, as yet unidentified animal, which researchers speculate may now be extinct.

A different kettle of fish

It also appears that, as in the case of flowering plants, hybridisation can fuel explosive radiations of novel animals. The best-known example is the case of the cichlids of Africa’s Great Lakes—particularly Lake Victoria, Lake Tanganyika and Lake Malawi. Great Lake cichlids are a group of thousands of closely related fish, famous for their panoply of shapes, sizes and colours (see picture). Each is adapted to a different depth and ecological niche.

     Gone fishing

Cichlids’ evolutionary history has long puzzled biologists. Lake Victoria, in particular, comes and goes with the climate. Its current instantiation is less than 15,000 years old. 

In evolutionary terms this is the blink of an eye, but in that time the lake’s cichlids have diversified into more than 500 species.

The reason is hybridisation. Using genetic analysis to place Lake Victoria’s cichlids within the broader cichlid family tree, researchers have discovered that they descend from a tryst between two distinct parental lineages, one that swam in the Congo and the other in the Nile.

The value of being such a genetic mosaic is apparent from the history of one of the best-studied cichlid genes, which encodes a protein called long-wave-sensitive opsin that is found in the retina of the eye. This protein determines the eye’s sensitivity to red light. That matters because red-light levels decline steeply in deeper water. 

Consequently, fish which live at different depths need eyes that are tuned differently from one another.

The cichlid lineage from the Congo had eyes which were optimised for clear, shallow water. Nile-lineage vision was more attuned to the deep and murky. Hybrids were able to chop and change these genetic variants to produce a range of sensitivities to light. 

This let them colonise the full depth of the water column in Lake Victoria as it developed. The new lake, for its part, offered the cichlids a host of empty ecological niches to fill. The result was a sudden and explosive process dubbed “combinatorial speciation”.

Elsewhere in the natural world, combinatorial speciation seems to have contributed to the striking diversity of Sporophila, a genus of 41 Neotropical songbirds, and of the munias, mannikins and silverbills of the genus Lonchura, a group of 31 estrildid finches that ranges across Africa and South-East Asia. 

Nor is it just in vertebrates that this phenomenon rears its head. Heliconius, a genus of 39 flamboyant New World butterflies, also owes its eye-catching diversity to combinatorial speciation.

Raining cats, dogs and bears

These findings muddy Darwin’s concept of speciation as a slow and gradual process. 

Biologists now know that in the right circumstances, and with the help of hybridisation, new species can emerge and consolidate themselves in a mere handful of generations. 

That is an important amendment to evolutionary theory.

It is nevertheless true that, for animals, hybrid speciation in its full form remains rare. It requires an unlikely congruence of factors to keep a new hybrid population reproductively isolated from both parental species. 

The survival of the Galápagos Big Bird lineage, for example, involved physical isolation from one and strong sexual selection against the other.

More commonly, an incipient hybrid population is reabsorbed by one or both parental species before it can properly establish itself. The result is a percolation of genes from one species to another, rather than a full hybrid. 

This is called introgressive hybridisation—or, simply, introgression. DNA analysis of a long list of closely related animals shows that this version of hybridisation is far more common than the full form. 

It may even be ubiquitous.

The North American grey wolf, for example, owes its gene for melanism—the deep black fur displayed by some wolves—to introgression from domesticated dogs brought 14,000 years ago from Asia by America’s first human settlers. 

In wolves that inhabit forests this gene has undergone strong positive selection, suggesting it is adaptive. The most obvious explanation is that melanism provides better camouflage in the stygian depths of North America’s woodlands. Alternatively, female wolves may simply prefer their males tall, dark and handsome.

Panthera—the genus to which most big cats belong—is yet more impressive in the scope of its introgressive entanglement. It has five members: lions, tigers, leopards, snow leopards and jaguars. These have long been known to interbreed successfully in captivity, yielding crosses called ligers (lion x tiger), jaglions (jaguar x lion) and so on. 

But recent analysis shows that this has also happened in the wild. Researchers have identified at least six past introgressive episodes in the genus, with every member involved in at least one of them.

The most promiscuous of the five appears to be the lion. Gene variants have percolated between lions and tigers, lions and snow leopards, and lions and jaguars. There is also evidence that at least some of this gene flow has been adaptive. 

Three lion genes incorporated into jaguar genomes are known to have been strongly selected for. Two of these are involved in vision—specifically, they help guide the development of the optic nerve.

Genetic analysis also reveals a long history of hybridisation between polar bears and grizzlies, the largest of their brown bear cousins. It is not yet clear whether this has had adaptive value—but it may soon have a chance to prove itself. 

As climate change warms the polar bear’s Arctic home, the species may have to adjust rapidly. A splash of grizzly, a group used to more temperate climes, might help that happen.

The best-studied case of introgression in animals is, though, closer to home than wolves, big cats and bears. It is looking back at you from the mirror. The most up-to-date evidence suggests that Homo sapiens arose more than 315,000 years ago from gene flow between a series of interlinked population groups spread across Africa. 

Whether these populations were different enough to be considered distinct species is still debated. In the grasslands of the African Pleistocene, however, these ancestral groups were not alone. Their world was interspersed with a menagerie of other hominins. And interspecies mating seems to have been rife.

My family and other hominins

Several members of this human menagerie appear to have descended from Homo heidelbergensis, a species that spread through eastern and southern Africa around 700,000 years ago before crossing the Middle East into Europe and Asia. 

This species—a possible ancestor of the progenitor groups of Homo sapiens—also gave rise to at least two others, the Neanderthals (Homo neanderthalensis) and the Denisovans (Homo denisova). 

The former survived in Europe until 28,000 years ago, while the latter, an Asiatic group, lasted until roughly 50,000 years ago.

Other hominin species around at the time emerged directly from Homo erectus, a more primitive creature that was also the ancestor of Homo heidelbergensis and which, a million years beforehand, had blazed a similar transcontinental expansionary path to that of heidelbergensis. 

The local descendants of erectus were largely displaced by heidelbergensis when it arrived. But some holdouts survived in corners of the Old World that heidelbergensis never reached. These included the islands of Flores in Indonesia and Luzon in the Philippines. 

It was here that diminutive Homo floresiensis and Homo luzonensis—the island “hobbits”—lasted, like the Denisovans, until 50,000 years ago. There were probably isolated descendants of even older cousins too. 

At least one is known, Homo naledi, which predated the emergence of Homo erectus and still roamed southern Africa around 230,000 years ago.

This grand hominin circus ultimately came to an abrupt end. The record in Africa is opaque. But in Europe, Asia and Oceania it is clear that the arrival of modern humans coincided with a great vanishing of local hominins. 

Whether through disease, competition for scarce resources or perhaps even genocide, a few thousand years of contact with Homo sapiens was enough to snuff out every other hominin species.

Ghosts from a distant past

Even a few millennia, though, proved enough for Homo sapiens to get to know its cousins intimately. The record of these romantic entanglements remains in the DNA of almost everyone alive today. In 2010 a team led by Svante Pääbo of the Max Planck Institute’s campus in Leipzig published the first draft sequence of the Neanderthal genome. 

This led to the discovery that stretches of Neanderthal DNA constitute 1-4% of the modern human genome in all populations outside sub-Saharan Africa. 

That is consistent with a string of hybridising liaisons in Europe, the Middle East and Central Asia from around 65,000 years ago.

Neanderthal inheritance helped Homo sapiens adapt to the demands of the environments of these unfamiliar places. There seems to have been strong selection, for example, in favour of Neanderthal genes related to skin and hair growth. 

These include bnc 2, a gene linked to skin pigment and freckling that is still present in two-thirds of Europeans. There also appears to have been selection for Neanderthal-derived genes that deal with pathogens. Some govern the immune system’s ability to detect bacterial infections. Others encode proteins which interact with viruses.

The Denisovans, and their contribution to Homo sapiens, were another of Dr Pääbo’s discoveries. In 2009 one of his team sequenced dna from a fossil finger bone excavated from Denisova cave in the Altai Mountains of Siberia. This bone turned out to belong to a previously unknown species that was then named after the cave it was found in. 

Physical specimens of this species remain rare. 

Examination of living people, however, reveals that stretches of Denisovan dna make up 3-6% of the genome of contemporary Papuans, Aboriginal Australians and Melanesians. Many Chinese and Japanese also carry Denisovan dna, albeit at lower rates.

As with Neanderthals, this inheritance has brought advantages. The Denisovan version of a gene called epas1 modulates production of red blood cells, which carry oxygen. 

This helps modern Tibetans to survive at high altitudes. Denisovan tbx 15 and wars 2 similarly help Inuit survive the harsh cold of the Arctic by regulating the amount of metabolic heat they produce.

We contain multitudes

That the Denisovans could lurk in modern human dna yet leave so little fossil trace has caused geneticists to wonder what other ghosts they might find. The genomes of sub-Saharan Africans, in particular, reveal evidence of at least one further entanglement. 

In 2012 a genomic analysis of members of the Baka, Hadza and Sandawe, three groups of people of ancient lineage, suggested an archaic introgression. In 2016 a deeper analysis focused on the Baka pinpointed this to within the past 30,000 years. This February, a study of members of two other groups, the Yoruba and Mende, confirmed that between 2% and 19% of their genomes can be traced to an unidentified archaic species. 

Whether this is the same as the one which has contributed to the Baka, Hadza and Sandawe is unclear, but it appears to have diverged from the line leading directly to Homo sapiens not long before the Neanderthals and Denisovans—an African Neanderthal, if you will.

The same genetic tools have revealed deeper ghosts, too. Denisovans show signs of hybridisation with a “superarchaic” lineage—perhaps Homo erectus itself. This makes up 1% of the species’ genome. 

About 15% of this superarchaic inheritance has, in turn, been passed on to modern humans. There is even evidence of a minute genetic contribution to African populations by a similarly superarchaic relative.

To be human, then, is to be a multispecies mongrel. As the example of the big cats in particular shows, though, Homo sapiens is not, in this, an exception. 

Hybridisation, once seen as a spear-carrier in evolution’s grand theatre, is rapidly becoming a star of the show. Meanwhile, Darwin’s idea of a simple, universal family tree is relegated to the wings.

In its place, some experts now prefer the idea of a tangled bush of interconnected branches. But this, too, is an imperfect comparison. 

A more fitting analogy is a frayed rope. Species are braided from individual strands. Where evolution proceeds in an orthodox Darwinian manner, braids unravel, strands split and new species result. But the rope does not fray neatly. 

Filaments of introgression criss-cross from braid to braid and, occasionally, two tangle to form a new braid altogether. This is a more complex conception of evolutionary history, but also a richer one. 

Few things in life are simple—why should life itself be?