FACT: Arctic sea ice cover is down by 3.2 million square kilometers — an expanse of water as big as India — since 2000.

 

THE NUMBERS: Carbon dioxide emissions intensity in 2023, in grams per $1 of GDP* –

Country	Value
Iran	54
China	42
Russia	36
Saudi Arabia	34
Canada	26
World Total	24
India	23
United States	19
Indonesia	17
Japan	16
European Union	10
United Kingdom	8
Sweden**	5
Ireland**	5
Singapore**	5
Switzerland**	5

Source: EDGAR, the EU’s “Emissions Database for Global Atmospheric Research”
** A 4-way tie for lowest emissions intensity rate among major economies.

 

WHAT THEY MEAN:

The Arctic sea ice expands and contracts like a lung, growing through autumn and winter to a late-March peak, then shrinking back over the summer to a mid-September “minimum.” The 2024 minimum is likely this week. Reviewing their 46 years of data, the National Snow and Ice Data Center in Colorado concludes that it will be the fourth-smallest on record:

“With the waning of sunlight, the pace of sea ice loss in the Arctic is slowing, and the seasonal minimum is expected in mid-September. While a new record low is highly unlikely, extent at the beginning of September is below many recent years. Both the northern and southern Northwest Passage routes have largely cleared of ice, as has the Northern Sea Route. … Antarctic ice extent is approaching its seasonal maximum and is near last year’s record low. Arctic sea ice extent as of August 31 was 4.55 million square kilometers (1.76 million square miles), fourth lowest in the 46-year passive microwave record for that date.”

NSDIC says Arctic ice cover has contracted by about 10% per decade since the millennium. The cumulative loss of ice, averaged over a full year, comes to 3.2 million square kilometers — a space of water about as big as India — and means ice coverage is down about 40% from the 8 million square kms typical of the 1980s.  The ice is also (a) thinner — now mostly 2 to 3 meters from top to seawater, which is about half the 5-meter average Norwegian polar explorer Fridtjof Nansen found on his first-ever Arctic ice transit in 1893 — and (b) younger.  In the 1980s, “old” ice four years old or more made up about a third of polar ice volume, and “new” ice less than one year old likewise a third; this decade, about 5% of ice is “old” and 70% “new.”

The diminished polar ice cap is an especially visible reminder that governments and industries have limited time to think and argue. Moving from ice to emissions, statistics from the EU’s EDGAR database — updated Thursday for 2023 emission totals, a day after the NSIDC published its 2024 Arctic minimum estimate — suggest three things. One, worldwide, the “battle” to contain climate change, if “battle” is the right metaphor, is being lost. Two, it’s being lost mostly in large middle-come countries – though carbon emissions from rich countries are now falling, those from China in particular, and also India and some other big countries, more than offset these drops; and three, the “battle” isn’t lost yet. Here’s a summary:

Totals: The number that counts for ice, forests, corals, northern mammals, sea life, arable land, and coastal community safety is the level of greenhouse gases in the atmosphere.  Here the annual human contribution continues to rise. In 2000, human carbon dioxide emissions came to 25.2 billion tons.  The 2023 total was 39.0 billion tons, meaning annual CO2 emissions are up 50% so far this century.  Alternatively, since 2020 (an unusually low year because of the COVID pandemic), carbon emissions are up by 2.9 billion tons, or 8%.

Emission by country: Within this big emissions total, smaller country-by-country shares (based on territorial output) have sharply changed. The “developed” world — the U.S. and Canada; the EU, the U.K., Switzerland, Norway, and Iceland; Japan, Taiwan, and Korea; Israel; Australia and New Zealand — has cut annual carbon dioxide emissions by 2.5 billion tons since 2000.  Though they are still far from “net zero,” they are trending down at an accelerating pace. Annual CO2 emissions from the rest of the world, though, are up 16 billion tons, including by 9.6 billion tons in China, 2.0 billion tons in India, and about 4.0 billion tons elsewhere. In sum, the growth of Chinese and middle-income country emissions has far outdistanced the “developed” countries’ reduction.

“Emissions intensity”: Against this overall growth of emissions, trends in emissions “intensity”– that is, the amount of carbon produced per dollar of output — offer some reason for guarded optimism.  The United States and Europe have not cut emissions by getting poorer but by growing more efficient.  In 2000, Americans produced 38 grams of carbon for each dollar of (real, inflation-adjusted) GDP, and EU countries 21 grams. By 2023, the U.S. was down to 19 grams per dollar, and the EU to 10, and the world’s most carbon-efficient economies suggest that there’s still a lot of room to improve. The U.K. and France are at 8 grams of carbon per dollar, Denmark 6 grams, and Sweden, Ireland, Singapore, and Switzerland set the world standard at 5 grams.

This positive trend isn’t unique to the wealthy world: Chinese emissions intensity is down 44% – from 73 grams per dollar in 2000 to a still-high 42 last year since 2000 — and India’s by 24%.  In only three of the top 20 emissions sources — Iran, Saudi Arabia, and Vietnam — has intensity grown since 2000. Here’s a table summarizing emissions from the largest country sources, including total CO2 emissions in 2023, change in this total since 2000, and change in emissions intensity since 2023, with positive trends colored green and negative ones red:

Country	2023 Emissions (billion tons)	Change 2000-2023	Emissions/GDP ratio change since 2000
World	39.0	+13.3 billion tons	-27%
China	13.3	+9.6	-44%
United States	4.7	-1.2	-50%
India	3.0	+2.0	-24%
EU	2.5	-0.9	-52%
Russia	2.0	+0.4	-36%
Japan	0.9	-0.3	-36%
Iran	0.8	+0.4	+10%
Indonesia	0.7	+0.4	-26%
All other	8.1	-0.2	-44%**

* Using EDGAR data released last week. Other estimates, such as those by IEA, differ slightly but not fundamentally.  

** Not available in EDGAR; PPI estimate using World Bank GDP data.

Final point: The Ireland/Singapore/Sweden/Switzerland 5-grams-per-dollar intensity standard is much better than any very large emissions source country has achieved.  But their diverse economic mix — Ireland and Singapore with lots of high-end manufacturing and big computer servers; Sweden, Switzerland, and Ireland with big farm and livestock sectors; Sweden with its six auto plants; Switzerland and Singapore as services and logistical centers — suggests that bigger countries should be able to match their record.  Had the world’s top 8 emissions sources done so, their emissions would have been 5.8 billion tons rather than 28 billion tons last year, and as a group, they would have been 80% of the way to net zero. So, that’s not an impossible goal, or even an unrealistic one.  The ice bulletins say clearly that the ‘battle’ isn’t now being won. But it isn’t yet lost.

FURTHER READING

Ice at the Pole –

The National Snow and Ice Data Center reports on this year’s Arctic ice minimum.

… and for a look at the mountains, the National Park Service on Glacier Park.

Emissions data sources – 

The European Union’s EDGAR (“Emissions Database for Global Atmospheric Research”) database, out last Thursday.

The International Energy Agency’s report for 2023, from last March.

Direct reports –

German scientist Markus Rex recounts the two-year “MOSAIC” mission, in which a specially equipped icebreaker sealed itself into the Arctic ice in 2022 north of Svalbard, and came out on the other side near Greenland two years later.

Fridtjof Nansen’s wooden Fram did the same thing in the first successful Arctic Ocean research and survey mission, from 1893 to 1896. The Fram bogged down (though it was later recovered); Nansen, after making the first estimates of Arctic ice thickness, ocean depth, and wildlife diversity, got to Greenland by sled and kayak.

Trade & policy –

The European Union’s Carbon Border Adjustment Mechanism.

The International Maritime Bureau on efforts to cut ship emissions (which, per EDGAR, are about 0.8 billion tons of carbon dioxide a year, roughly equivalent to seventh-ranking Iran).

Policy and research from top performers and one of the most threatened countries –

Singapore, with Asia’s most efficient emissions intensity, looks at climate change impact at home.

Ireland’s Environmental Protection Agency.

Sweden’s Bolin Institute at the University of Stockholm.

Switzerland’s climate strategy.

And Bangladesh’s Ministry of Environment, Forests, and Climate Change.

FACT: Trump-era tariffs raised prices but did not ‘bring manufacturing back’.

 

THE NUMBERS: Manufacturing share of U.S. GDP –

Year	Percentage
2023	10.3%
2020	10.1%
2016	10.9%

 

WHAT THEY MEAN:

Here’s Vice Presidential candidate Senator J.D. Vance, pitching higher tariffs in late July and saying no price for toasters would be too high. And there was the same Mr. Vance, a week earlier, denouncing high prices and their toll on middle-class family budgets: “Many of the people that I grew up with can’t afford to pay more for groceries, more for gas, more for rent.”

Visionary poets like Walt Whitman can get away with this sort of thing. (Song of Myself, verse 51: “Do I contradict myself? Very well, I contradict myself. (I am large, I contain multitudes.)”.) Politicians struggle. Challenged in a TV interview a few days later, Vance tries hard to have it both ways, with a claim about earlier Trump-era tariffs:

“[M]anufacturing came back and prices went down for American citizens. They went up for the Chinese but went down for our people.”

Is he right? In fact, manufacturing growth slowed down while prices went up. The two main Trump-era tariffs went onto steel and aluminum in March 2018 (“Section 232”), and most Chinese-made goods (“Section 301”) — though toasters were exempted — in three pulses from September 2018 to mid-2019.  Overall, this raised average “trade-weighted” U.S. tariff rates from 1.4% to 3.0% as of 2019. Since then, the figure has dropped back to 2.4%, as some purchasing shifted from China to Vietnam and other countries.  Here’s the price and output data afterward:

Prices: The U.S. International Trade Commission’s 2023 report on the Trump-era tariffs is the standard source here. It finds prices up for families buying consumer goods and also up for manufacturers, construction firms, and others buying metal.  The representative quotes:

Steel and aluminum (pg. 124): “The increase in tariffs on steel and aluminum imports increased the relative price of imports and led consumers of steel and aluminum to increase sourcing from domestic suppliers. This increase in demand for domestic production of steel and aluminum resulted in increases in the price of domestically produced steel and aluminum and the quantity of domestic steel and aluminum production in these industries. However, the higher prices of steel and aluminum translated into higher costs of production inputs for downstream industries. This effect negatively impacted the downstream industries that purchase steel and aluminum because costs increase per unit of production.”

Chinese goods (pg. 145): “[T]he tariffs did not have a significant impact on the price received by Chinese exporters. On the other hand, the elasticity of the importer price with respect to the tariffs is close to one, indicating that importer prices rose about 1 to 1 in response to the tariff increase. This is consistent with the recent work of Amiti et al. (2019), Fajgelbaum et al. (2020), Carvallo et al. (2021), and Jiao et al. (2022), who also largely estimate full pass-through of recent tariff actions from exporters to importers.”

Academic literature, along with reviews by Federal Reserve staff economists, concurs.  These generally converge on a finding that the tariffs raised overall U.S. prices in a range of 0.3% to 0.5%, with the highest estimate a bit above 1.0%.

Output: With respect to manufacturing trends afterward, the ITC’s report concludes that (as of 2021) the steel and aluminum tariffs had raised the two metals’ output by about $2.2 billion, and shrunk the output of auto parts, machinery, tools, and other metal-using manufacturers (including household appliances such as toasters) by about $3.5 billion. On net, therefore, an overall slight shrinking of U.S. manufacturing. (They didn’t do a similar estimate of the China tariffs’ impact on consumers and firms buying inputs.)

More broadly, the post-tariff trend has been somewhat slower growth in manufacturing output and employment, though not an overall contraction. The Commerce Department’s Bureau of Economic Analysis, which calculates U.S. GDP stats, reports that in 2017, manufacturing made up $2.1 trillion of a $19.6 trillion American economy — that is, industries making planes, cars, semiconductor chips, frozen meat, refined petroleum, medicines, plastics, etc. accounted for 10.9% of overall output just before the tariffs. By 2019, their $2.2 trillion output was 10.5% of a $2.2 trillion economy; after a sharp drop to 10.0% in 2020 during the COVID pandemic, it rebounded to 10.3% in 2023 ($2.29 trillion of a $22.38 trillion economy). So manufacturing continued to grow, but makes up a somewhat smaller part of the U.S. economy than before the Trump-era tariffs.

Alternatively, in terms of “real,” inflation-adjusted growth, U.S. manufacturing grew by $33.5 billion on average per year during the Obama presidency, measuring from the financial crisis low of 2009 to 2017.  Since 2018 it has grown by $30.2 billion per year, again slightly more slowly after the tariffs than before.

Employment: The employment story is similar – a slowdown in net manufacturing job creation, and a somewhat smaller share of total employment. According to the Bureau of Labor Statistics, in July of 2017, manufacturers employed 12.5 million of 146.8 working Americans, or 8.5% of all jobs.  As of July 2024, the figure is 13.0 million of 158.7 million jobs, or 8.2%. Alternatively, again looking back to the Obama era, from the financial crisis low in mid-2009 to mid-2017, manufacturers added a net of 795,000 jobs. This was an average net gain of 99,500 jobs per year. Since mid-2017, they’ve added another 480,000, for 68,600 jobs per year. Or, in terms of wages, BLS’ stats find manufacturing workers earning about 27 cents more per hour than the national average in 2017 (a 1.0% advantage), and 29 cents per hour less (a 1.0% disadvantage) by 2021.

So: A lot goes into these big numbers – economic shocks and booms, career choices of young workers, innovation and adaptation of new technologies, etc. But the overall data from BEA and BLS make ITC’s conclusion that the 2018/19 tariffs raised prices and slightly shrank U.S. manufacturing look pretty strong. Vance’s backward-looking claims of ‘lower prices’ and ‘coming back’ don’t hold up. You can, however, be pretty confident about his promise that with an added tariff of 10% or 20%, toasters – along with other home appliances, groceries, gas, TV sets, refrigerators, cars, medicine, toothpaste, and consumer goods generally – will cost more.

FURTHER READING

Rhetoric –

Whitman’s Song of Myself (see verse 51).

Sen. Vance on toaster prices being too low.

… but prices are also too high.

… and for a longer tariffs-and-prices exchange, an August 25th “Meet the Press” interview transcript.

Data –

USITC’s March 2023 report on Trump-era tariffs (with modeling and evaluation through year 2021).

GDP-by-industry data from BEA.

… and employment and wage stats from BLS (in “Employment, Hours, and Earnings”).

And some more on toasters –

Vance’s toaster-price rhetoric appears to have contaminated the kitchen-appliance Internet with a moldy spread of opportunistic ‘made-in-USA toaster’ false positives. Discarding these, our search finds that no small home pop-up toasters appear to be made in the United States at the moment.  U.S. factories do, however, produce kitchen and other home appliances – according to BLS, about 61,500 people work in appliance production – and several firms (e.g. Tennessee-based Holman Star) make big conveyor-type toasters for restaurants and hotels.  They sell for $1,000 and up. Alternatively, the UK’s Dualit makes its “Classic” line by hand in Sussex, starting at £170 (=$222).

BLS on home-appliance industry employment and pay.

And tariffs appear to be making it harder, not easier, to make these things here. From the political right, National Review’s Dominic Pino reports on toasters and American kitchen appliance-makers’ unhappy experience with Trump-era metals tariffs.

FACT: World infant mortality rate cut by half since 2000; in the U.S., by only 22%.

THE NUMBERS: Infant mortality rates, 2000-2022* –

Country	2000	2022	Decline
Sierra Leone (highest rate 2022)	138.3	76.0	-45%
Estonia (lowest rate 2022)	8.7	1.5	-83%
Least-developed countries	108.6	42.3	-61%
World	53.0	27.9	-47%
High-income countries	6.9	4.1	-41%
United States	7.1	5.6	-22%

Deaths in the first year of life, per thousand live births. World Bank for world and regions; CDC’s most recent release for the U.S. rate in 2022.  Note that the CDC has recalculated the U.S.’ 2000 figure to 6.9 per 1000.

WHAT THEY MEAN:

“Natalism,” the view that governments should encourage people to have children, can come in modest and supportive forms: child tax credits, cash subsidies for child care, and so on.  Few are extremely successful.  More exotic approaches — e.g. Vice Presidential candidate Vance’s haranguing of childless couples as in some way letting down the country — aren’t at all likely to do better.  An alternative suggestion: improve the protection of children once they’re born.  Here a remarkable 21st-century international success — a sharp drop in infant mortality almost everywhere, fastest on average in poorer countries but also clear in Europe, Australia, and Japan — highlights an area where Americans should do better. The background and the contrast:

20th-century success:  In the very long view, American infant mortality has declined drastically — by 99.4% — since 1915, when the Labor Department’s newly formed Children’s Bureau first estimated a national infant mortality rate. Their statisticians found 99.9 deaths in the first year of life per every 1,000 births that year, for a 10% infant mortality rate.  Earlier rates were likely even higher — Michigan’s health department, for example, reported 15.7% state infant mortality in 1900 — and the Bureau believed 30% of all deaths in the U.S. were of children five years or younger. To put these figures in perspective, the death rate for soldiers in the Union Army was 13.4%; and according to World Bank tables, the highest national infant mortality rates today are Sierra Leone’s 76 per 1,000 and the Central African Republic’s 74. Or, to move from data to family life, early childhood death was common, and neither power and wealth, nor education, nor medical education seem to have been defenses: three of Abraham and Mary Lincoln’s four children died young, as did four of Karl and Jenny Marx’s seven, three of Louis and Marie Pasteur’s five, and two of Charles and Emma Darwin’s ten.

The causes were fairly simple: (a) bad sanitation, especially contaminated milk and water; (b) no vaccines, anti-inflammatory medicines, or antibiotics; and (c) poor or non-existent primary health care. Over the next decades, with public health and sanitation laws, new medicines, vaccination campaigns, Medicaid insurance for the poor, and other policy innovations, the figures collected by the Children’s Bureau and its successors at the Centers for Disease Control steadily improved: 60 deaths per thousand births by 1930, 40 by 1940, 26 by 1960, 10 by 1980, and 7.1 (since recalculated to 6.9) as CDC published its “Achievements in Public Health ” retrospective in 2000.

But more recently: Placed against this 20th-century achievement, the U.S.’ 21st-century record looks mediocre at best. CDC’s most recent calculation of the national infant mortality rate (for 2022, out in late July) is 5.6 deaths per 1,000 births.  This is 22% below the rate of 2000 and actually up from 5.4 deaths in 2021 – the largest jump in 20 years.

Meanwhile, the rest of the world has cut infant mortality by nearly half.  It’s tempting to put the relatively poor U.S. performance in a good light by noting that the high worldwide rate reflects remarkable successes in lower-income countries, and/or assuming that public health policy might have a diminishing rate of return, with infant mortality reduction slowing naturally as rates fall and at some point stopping.

The first of these points is true. Most of the low- and middle-income world has cut infant and childhood deaths deeply and fast since 2000. Uzbekistan, for example, has cut infant mortality by 77% since 2000, Cambodia by 76%, Armenia and Morocco by 66%, Bangladesh by 62%, Ethiopia 61%, Egypt 59%, El Salvador 57%, Indonesia 56%, Honduras 55%, Timor-Leste 53%, and Ghana 51%. But the second isn’t.  The U.S.’ rate was near the ‘developed’-world average in 2000, is now clearly higher, and has fallen more slowly than most. As an example, Japan’s 3.2 per 1000 rate was the world’s lowest in 2000, and has fallen at exactly the world average rate to 1.7 per 1000.  A representative table of large and advanced economies:

Country	2022 Infant Mortality Rate	Drop since 2000
China	4.8	-84%
Australia	3.2	-63%
Korea	2.5	-63%
Ireland	2.7	-60%
India	25.5	-62%
Italy	2.2	-53%
World	27.9	-47%
Japan	1.7	-48%
Spain	2.5	-43%
United Kingdom	3.6	-39%
Germany	3.0	-32%
United States	5.6	-22%
France	3.3	-19%
Canada	4.3	-19%

What explains this?  And what might be done?

One explanation comes from internal disparities: differing experiences by geography, by race, and ethnicity, and differing rates of change over time. Some comparisons:

* State and region: By state, Massachusetts has the lowest U.S. infant mortality rate at 3.2 per 100 births; Mississippi’s 9.1 per thousand is the highest. By region, rates in New England, the mid-Atlantic, and the West Coast are well below the national averages and within a standard rich-country range. Rates in the deep South and Vance’s Ohio-Indiana-West Virginia region are typically above 7 per thousand.

* Race and ethnicity: Infant mortality rates range from 3.7 per thousand in Asian American families through 4.4 in non-Hispanic white families, 4.8 in Hispanic families, 7.5 in Native American, and 10.6 in African American families.

* Change over time: By race and ethnicity, declines over time are not vastly different.  Regionally, however, some parts of the country have cut infant mortality noticeably faster than others. The best state record since 2000 is the District of Columbia’s 60% reduction, followed by drops of 45% in New Jersey and North Dakota, 39% in Rhode Island, 36% in Massachusetts, and 35% in New Hampshire.  Recent studies suggest a couple of commonalities in the latter group.  One is non-participation in the Affordable Care Act’s Medicaid expansion, which improved access to prenatal and maternal care for lower-income moms.  Another is the imposition of abortion restrictions since the Supreme Court overturned Roe v. Wade in 2022, often vaguely written and applicable to many different situations, which may be both keeping patients from seeking care and deterring providers from working in the relevant areas.
Perhaps related to both of these themes is an ominous trend in rural health care: maternity care seems to be getting steadily harder to find, and both expectant mothers and infants therefore face rising risk. Health consultancy Chartis summarizes:

“More than 400 maternity programs closed nationwide between 2006 and 2020. In rural communities, the disappearance of OB services has been particularly impactful. Between 2011 and 2021, 267 rural hospitals closed OB services, representing 25% of all rural OB units in the United States.”

To put these figures in context, there are 2,168 maternity hospitals in the United States.  So at least in remote areas, health care options are fewer and further away than they were a generation ago. And bringing this from policy and data to family life, the CDC reports 20,577 infant deaths in 2022. Had the U.S. achieved Japan’s world-standard 1.7 per thousand infant mortality rate, 14,300 of them would have lived. The EU’s 3.1 per thousand rate — which is certainly possible, since it’s not far away from the New England average — would have kept 9,200 alive.

So: For those wanting more American children, Vance’s “browbeating” approach is pretty certainly useless, and more prosaic things — child tax credits and child care support, better access to prenatal and perinatal care, primary health in general, “keeping clinics open in places where they’re closing” — are good ideas. We can, really, do a lot better.

Source: OECD

FURTHER READING

U.S. –

CDC has infant mortality rates, overall and by race and ethnicity.

… a state-by-state map.

Chartis on closures of rural maternity wards.

NIH on Medicaid expansion.

… and a study looking at Medicaid expansion, post-Dobbs abortion restrictions, and changes in infant mortality rates.

World perspective –

The World Bank has rates for 1990, 2000, and 2014 through 2022 across all countries and for regions and income groups.

… and a sample of its data:

Country	2000	2022	Decline
China	29.9	4.8	-84%
Estonia	8.7	1.5	-82%
Uzbekistan	51.4	11.9	-77%
Mongolia	48.4	11.5	-76%
Turkey	30.9	8.5	-73%
Cambodia	79.0	22.3	-72%
Armenia	27.0	9.2	-66%
Australia	5.1	3.2	-63%
Bangladesh	63.0	24.1	-62%
Least-developed countries	108.6	42.3	-61%
Ethiopia	87.4	33.9	-61%
El Salvador	24.6	10.5	-57%
Indonesia	40.9	18.1	-56%
Honduras	30.5	13.8	-55%
Ireland	6.0	2.7	-55%
Italy	4.7	2.2	-53%
Timor-Leste	87.3	41.5	-53%
Mexico	23.6	11.5	-51%
Japan	3.3	1.7	-48%
European Union	5.9	3.1	-47%
World	53.0	27.9	-47%
Jordan	22.6	12.2	-46%
Sierra Leone	138.6	76.0	-45%
High-income countries	6.9	4.1	-41%
United States	7.1	5.6	-24%
Canada	5.3	4.3	-19%
Jamaica	17.5	16.1	-8%

 

Success stories –

In 2000, the capital’s infant mortality rate, at 13.5 deaths per 1,000 births, was nearly twice the national average. As of 2022, it was 5.5, down by 60% and equal to the national average. Mayor Bowser’s health division, with links to perinatal and infant health services and reporting.

UNICEF looks at newborn health care in Uzbekistan.