Oceanography | Vol.30, No.2
and the NSF-OCE budget briefly grew
faster than inflation for the first time since
the early 1970s. Programs were launched
with great optimism. However, the actual
budget increases just got NSF-OCE fund-
ing levels out of a hole and back up the
level that tracked GDP growth.
Following the September 11, 2001, ter-
rorist attacks, a growing percentage of
federal funds were shifted to defense and
homeland security. Since 2004, inflation-
corrected purchasing power for ocean
sciences has declined steadily. If the OMB
2018 budget is enacted, NSF-OCE will
have been reduced to the purchasing
power it had in 1972 and 1992.
NSF details the implications of the
White House budget here: https://www.
nsf.gov/about/budget/fy2018/index.jsp;
the proposed cut for NSF-OCE is about
10.2% (9.8% cut in disciplinary and inter-
disciplinary science). In the absence of
reallocations of funding priorities within
NSF, if the OMB 2018 budget becomes
law, ocean infrastructure purchasing
power will be up about 20% relative to
2003 (due to cost inflation of fixed facil-
ities initiated when budgets were rising),
but science investigator funding (the
normal proposals we all write to do our
work) will be down 49%. We will have
lost nearly half the NSF science program
in oceanography over a 15-year period2.
Is this true of NSF as a whole? Yes
and no.
In Figure 1b, the red and blue sym-
bols are as before, but now for the total
NSF budget rather than just for OCE sci-
ence. As before, the dashed curves are
dollar amounts for a constant fraction of
GDP (here 0.04%, the average from 1970
to 2016). Figure 1b makes it clear that
until now, total NSF funding has more or
less tracked GDP growth since 1970. The
proposed cuts in the OMB 2018 budget
are not the first time NSF has seen cuts,
but they are severe and abrupt relative
to past history.
So why did NSF-OCE funding not
keep up with the overall NSF budget?
In Figure 1c, the blue curve is the NSF-
OCE funding as a percentage of GDP. In
the 1970s, NSF-OCE was allocated about
0.004% of GDP. That percentage dropped
precipitously in the late 1970s, is now less
than 0.002% of GDP, and is projected to
fall further to about 0.0015%. In contrast,
total NSF funding (the green curve) has
oscillated but overall has stayed relatively
constant as a fraction of GDP, recovering
after a dip during the Carter and Reagan
Administrations. The proposed FY2018
budget imposes on NSF a drop rela-
tive to GDP that is equivalent to those of
the Reagan years.
To make this even clearer, Figure 1d
shows what happened to OCE within
NSF. The percentage of NSF funds dedi-
cated to ocean sciences in the 1970s and
early 1980s ranged between about 8% and
9% of the total NSF effort. The sea change
for ocean sciences occurred even earlier
than the visible loss of dollars in the 1990s.
Ocean sciences appears to have dropped
in priority at NSF (i.e., as a percentage of
NSF’s budget) starting in the mid-1980s,
when NSF changed course to emphasize
investment in programs the agency con-
sidered most directly related to economic
competitiveness such as engineering and
computer science (Bloch, 1985). Since
that time, NSF has been gradually reduc-
ing its fractional commitment to ocean
sciences. OCE is now under 5% of NSF’s
overall effort.
Readers of Oceanography almost cer-
tainly share my belief that ocean sci-
ence is needed now more than ever, both
in the United States and globally. The
ocean remains the least explored part of
1 Data reported here are for funds committed by the National Science Foundation Ocean Sciences Division (NSF-OCE). The data source for the interval 2000–2014 comes
from NSF as part of the “Sea Change” report (NRC, 2015). Older data are gleaned from tables in the National Science Board Science and Engineering Indicators. Actual val-
ues from 2015 to 2017 are from the federal record. Projections are based on the White House proposed budget for FY2018. The numbers here reflect science operations
and activities within NSF-OCE; they exclude major infrastructure projects (so called MREFC funds) such as ship construction and the one-time, 2008 funds associated with
ARRA (the American Recovery and Reinvestment Act). Some additional ocean sciences research occurs in other divisions, for example in NSF’s Office of Polar Programs
(OPP), and those funds are not included here.
A caveat in this analysis is that alternate databases on federal programs exist by subject area (https://ncsesdata.nsf.gov/webcaspar), and in some cases give different val-
ues. Discrepancies appears to reflect the fact that fields specified in the ncsesdata product do not conform with NSF divisions, so a judgment must have been made about
how to translate NSF program data into ncsesdata categories. Inspection of the ncsesdata database reveals large and implausible oscillations in division budgets on the
order of $100 million; for example, in a single year (1991) the uncorrected oceanography budgets appear to shift down by $100 million while at the same time the combina-
tion of geology and environmental sciences budgets shift up by $100 million. The opposite shift occurs in 1996. Because of these discrepancies, the data provided directly
by NSF-OCE was used, gleaned mostly from biennial Science and Engineering Indicators reports. Categorizing these values correctly involved some decisions to avoid the
problem of funds jumping between pigeonholes. NSF staff kindly checked the estimates I made and agreed that they were reasonable. It should be noted, however, that
older values in the NSF database are not as complete or detailed as more recent values, so some errors may remain. It may no longer be possible to check the details in
the older data.
Inflation corrections were based on the US Consumer Price Index (https://inflationdata.com/Inflation/Consumer_Price_Index). Data on US Gross Domestic Product from
1970 to 2016 came from the US Bureau of Economic Analysis (https://www.bea.gov/national/index.htm). Projections from 2017 to 2020 are from the International Monetary
Fund IMF Projection (https://knoema.com/qhswwkc/us-gdp-growth-forecast-2015-2019-and-up-to-2060-data-and-charts).
There was no attempt here to evaluate dollars available per scientist requesting funds. Anecdotally, however, it appears that the pool of scientists working in oceanogra-
phy has grown substantially over the past several decades. For example, TOS was founded in 1978 with a few hundred members, and has grown by a factor of 10. Similarly,
membership in the American Geophysical Union (AGU) in 1980 was about 13,000, and now is over 60,000 (of course, not all AGU members are ocean scientists). If this popu-
lation growth in ocean sciences is correct, it implies less dollars of funding per scientist even if the amount of funding is stable or increasing, depending on the relative rates
of change in funding and population of scientists.
2 Calculation is as follows: in FY 2003, the NSF-OCE science budget was $301.47 million, and of that $114.69 million was for infrastructure support, and $186.77 million was for
science and related activities (NRC, 2015). Translated into inflation corrected 2016-equivalent dollars, these values would be $369.66, $150.91, and $245.75 million, respec-
tively. The values proposed in the FY2018 White House budget are $323.02, $190.77, and $132.25 million (constant dollars) or $307.49, $181.60, and $125.89 million (in
2016-equivalent dollars). Thus, in terms of purchasing power, the fractional changes from 2003 to the proposed FY2018 are a loss of 23% (total NSF-OCE) with a rise of 20%
for infrastructure support, and a loss of 49% in infrastructure and related activities. For comparison, the change in total NSF budget over the same period (in 2016-equivalent
dollars) is a loss of 10% in purchasing power, almost entirely due to the cuts proposed for 2018.