• Open Access

Conserving imperiled species: a comparison of the IUCN Red List and U.S. Endangered Species Act

Authors

Errata

This article is corrected by:

  1. Errata: Corrigendum for Harris et al. 2011, http://onlinelibrary.wiley.com/
    doi/10.1111/j.1755-263X.2011.00205.x/abstract
    Volume 5, Issue 2, 157, Article first published online: 9 April 2012

  • Editor Phillip Levin

J. Berton C. Harris, Environment Institute and School of Earth and Environmental Sciences, University of Adelaide, SA 5005, Australia. Tel: +61 8 8303 5254; fax: +61 8 8303 4347. E-mail: bert.harris@adelaide.edu.au, aramidopsis@gmail.com

Abstract

The United States conserves imperiled species with the Endangered Species Act (ESA). No studies have evaluated the ESA's coverage of species on the International Union for Conservation of Nature (IUCN) Red List, which is an accepted standard for imperiled species classification. We assessed the ESA's coverage of IUCN-listed birds, mammals, amphibians, gastropods, crustaceans, and insects, and studied the listing histories of three bird species and Pacific salmonids in more detail. We found that 40.3% of IUCN-listed U.S. birds are not listed by the ESA, and most other groups are underrecognized by >80%. Species with higher IUCN threat levels are more frequently recognized by the ESA. Our avian case studies highlight differences in the objectives, constraints, and listing protocols of the two institutions, and the salmonids example shows an alternative situation where agencies were effective in evaluating and listing multiple (related) species. Vague definitions of endangered and threatened, an inadequate ESA budget, and the existence of the warranted but precluded category likely contribute to the classification gap we observed.

Introduction

Imperiled species lists have a variety of important uses that include classifying species’ conservation status, setting conservation priorities, and directing management (de Grammont & Cuarón 2006). While some imperiled species lists have been criticized because of their qualitative nature and application to multiple objectives (Possingham et al. 2002), the lists are firmly established as valuable tools for biological conservation (Lamoreux et al. 2003; Miller et al. 2007; Mace et al. 2008). The IUCN Red List is the most widely used global imperiled species list (e.g., Rodrigues et al. 2006; Schipper et al. 2008; BLI 2010), and its classifications are correlated with other leading systems such as NatureServe (O’Grady et al. 2004; Regan et al. 2005). The Red List classifies species as imperiled (Critically Endangered, Endangered, or Vulnerable), not imperiled (Near Threatened or Least Concern), extinct (Extinct, Extinct in the Wild), or Data Deficient (IUCN 2001, 2009). If species meet quantitative thresholds of any of the following criteria they will be added to the Red List: (1) decline in population size, (2) small geographic range, (3) small population size plus decline, (4) very small population size, or (5) quantitative analysis. For example, if a species had an estimated population size of <2,500 mature individuals, and had undergone a continuing decline of ≥20% over the last 5 years, it would be classified as Endangered. The IUCN Red List, like any categorical imperiled species classification, must make normative decisions that include risk tolerance in the designation of category boundaries; see IUCN (2001) for more details, and Mace et al. (2008) for the development and justification of Red List methods.

In addition to global imperiled species lists, many countries produce national red lists (local or regional imperiled species lists). These lists serve five major functions: (1) classifying the status of species at the local level where they are usually managed, (2) evaluating locally imperiled species and imperiled subspecies, (3) informing local conservation prioritization, (4) providing data to the global Red List, especially for species not yet evaluated by the IUCN, and (5) in some cases, legally protecting species (Miller et al. 2007; Rodríguez 2008; Zamin et al. 2010). See http://www.nationalredlist.org/ for an up-to-date listing of countries with national red lists and the methods they employ.

One of the most prominent and legislatively important national red lists is the U.S. Endangered Species Act (ESA). The ESA, passed in 1973 and administered by the U.S. Fish and Wildlife Service (USFWS) and National Marine Fisheries Service (NMFS), classifies an at-risk species (including subspecies and distinct populations) as endangered if it is “in danger of extinction throughout all or a significant portion of its range” or threatened if it is “likely to become endangered in the foreseeable future throughout all or a significant portion of its range” (USFWS 2009a; Figure S1; see supporting information). The USFWS is responsible for listing terrestrial and some marine species, while the NMFS lists marine species. Once a species is listed, the agencies work toward legally prohibiting “take” (killing, capturing, etc.), protecting critical habitat, and developing and implementing recovery plans for listed species (Schwartz 2008). Take of endangered animals is unconditionally prohibited, but for plants, only if they are on federal land. The agencies may develop a 4(d) rule to apply take prohibitions to threatened species. Designation of critical habitat and implementation of recovery plans are complicated processes that are not automatically applied by the USFWS (Schwartz 2008). The ESA has the power to stop development that will impact imperiled species. Hence there are more consequences and political obstacles to listing species under the ESA compared to lists that are not legally binding.

In short, the ESA is arguably the world's most effective biodiversity protection law. The act has succeeded in improving the conservation status of most listed species over time, and may have prevented 227 extinctions (Taylor et al. 2005; Schwartz 2008). Nonetheless, the U.S. government's implementation of the ESA has been problematic, including poor coverage of imperiled species (Wilcove & Master 2005), inadequate funding (Miller et al. 2002; Stokstad 2005), and political intervention (Ando 1999; Greenwald et al. 2006; Stokstad 2007). Despite the existence of the ESA, an extinction crisis continues in the United States (Elphick et al. 2010; Figure S2). For instance, 29 species and 13 subspecies went extinct while being considered for listing from 1973–1995 (Suckling et al. 2004). Most of these species already had very small population sizes when listing was proposed (sensuMcMillan & Wilcove 1994), but several species, such as Curtus's pearly mussel (Pleurobema curtum), likely could have been conserved had they been listed rapidly (Suckling et al. 2004).

Studies have analyzed the ESA's coverage of species on the NatureServe list, a leading classification of imperiled species in the United States (http://www.natureserve.org; Stokstad 2005; Wilcove & Master 2005; Greenwald et al. 2006), but, to our knowledge, no previous work has evaluated the ESA's coverage of IUCN-listed species. In the most comprehensive NatureServe comparison, Wilcove & Master (2005) investigated the ESA's coverage of plants, fungi, and animals considered imperiled on NatureServe's (2005) list. Wilcove & Master (2005) estimated that at least 90% of the country's imperiled species are not covered by the ESA. Given that the Red List is becoming the benchmark for global imperiled species classifications (e.g., Mace et al. 2008), an evaluation of the ESA's coverage of IUCN-listed species is needed. We refined previous work by focusing on birds, which are one of the best-known animal groups, and for which classification patterns might approximate a best case scenario. Then we looked in detail at three IUCN-listed birds that are not ESA-listed and, more generally, Pacific salmonids as case studies of classification under the ESA. We also compared classifications of insects, crustaceans, gastropods, amphibians, and mammals to evaluate if similar patterns existed to the previous NatureServe comparisons. Considering Wilcove & Master's (2005) results, we hypothesized that many U.S. IUCN-listed species would not be recognized by the ESA, and that poorly studied and lower-risk species (Vulnerable compared to Critically Endangered) would more likely be overlooked.

Methods

Our evaluation of the ESA's coverage of IUCN-listed species was not intended to evaluate extinction risk, but to provide a general indication of the breadth of coverage of the ESA compared to the Red List. The Red List—based on proxy measures of risk—is imperfect, but it is the most widely used, and among the most encompassing systems for global and national red lists (Lamoreux et al. 2003; de Grammont & Cuarón 2006; Rodrigues et al. 2006; Miller et al. 2007; Mace et al. 2008).

We compared classifications for all IUCN-listed birds known to be resident or fairly common visitors in the United States including Hawaii and Alaska (Pyle 2002; Dunn & Alderfer 2006). IUCN classification data came from BirdLife International's website (BLI 2010); ESA classifications came from the ESA website (USFWS 2009b). We followed the taxonomy of Chesser et al. (2010). If the ESA listed a single subspecies or a single population of an IUCN-listed species we considered the species to be covered by the ESA. We also collated data on Extinct, Extinct in the Wild, and Possibly Extinct birds (BLI 2010) and plotted these over time. Our extinction data were collected independently but are complimentary to Elphick et al.'s (2010) analysis that focused on estimating extinction dates.

For the case studies we examined IUCN-listed birds in Table 1 that were evaluated by the ESA, yet still not ESA-listed. We selected three species with adequate conservation status information and well-documented listing histories: Kittlitz's murrelet (Brachyramphus brevirostris), ashy storm-petrel (Oceanodroma homochroa), and cerulean warbler (Dendroica cerulea). We reviewed the peer-reviewed and gray literature for each species to examine the species’ conservation status and IUCN and ESA listing history. While all three species have large or relatively large ranges, each has undergone population declines and been listed as imperiled by the IUCN since 2004. Given that these species were not selected randomly, we do not mean to imply that their cases can be generalized to all imperiled birds in the United States; rather, the case studies are examples of what can happen when declining, IUCN-listed species are considered for ESA listing. We also present the case of Pacific salmonids (Salmonidae: Oncorhynchus) as an example where U.S. agencies were successful at evaluating and listing multiple species proactively.

Table 1.  Endangered Species Act status (endangered (E), threatened (T), or not listed) of IUCN-listed extant and possibly extinct birds in the United States organized by IUCN category. Twenty-five of the 62 IUCN-listed imperiled birds in the United States are not listed by the Endangered Species Act (IUCN 2009; USFWS 2009b; BLI 2010)
Species and IUCN classificationESA classification
  1. aNot endemic to the United States; bPossibly extinct (IUCN 2009); cAttwater's race (Tympanuchus cupido attwateri).

Critically endangered
 Laysan duck (Anas laysanensis)  E
 California condor (Gymnogyps californianus)  E
 Eskimo curlew (Numenius borealis)a, b  E
 Kittlitz's murrelet (Brachyramphus brevirostris)a  Not listed
 ivory-billed woodpecker (Campephilus principalis)a  E
 millerbird (Acrocephalus familiaris)  E
 olomao (Myadestes lanaiensis)b  E
 puaiohi (Myadestes palmeri)  E
 nihoa finch (Telespiza ultima)  E
 ou (Psittirostra psittacea)b  E
 palila (Loxioides bailleui)  E
 Maui parrotbill (Pseudonestor xanthophrys)  E
 nukupuu (Hemignathus lucidus)b  E
 akikiki (Oreomystis bairdi)  E
 Oahu alauahio (Paroreomyza maculata)b  E
 akekee (Loxops caeruleirostris)  E
 akohekohe (Palmeria dolei)  E
 poo-uli (Melamprosops phaeosoma)b  E
 Bachman's warbler (Vermivora bachmanii)a, b  E
Endangered
 Gunnison sage-grouse (Centrocercus minimus)  Not listed
 Hawaiian duck (Anas wyvilliana)  E
 black-footed albatross (Phoebastria nigripes)a  Not listed
 black-capped petrel (Pterodroma hasitata)a  Not listed
 Newell's shearwater (Puffinus newelli)  T
 ashy storm-petrel (Oceanodroma homochroa)a  Not listed
 whooping crane (Grus americana)a  E
 marbled murrelet (Brachyramphus marmoratus)a  T
 akiapolaau (Hemignathus munroi)  E
 Hawaii creeper (Oreomystis mana)  E
 Maui alauahio (Paroreomyza montana)  Not listed
 akepa (Loxops coccineus)  E
 golden-cheeked warbler (Dendroica chrysoparia)a  E
 tricolored blackbird (Agelaius tricolor)a  Not listed
Vulnerable
 Hawaiian goose (Branta sandvicensis)  E
 Steller's eider (Polysticta stelleri)a  T
 greater prairie-chicken (Tympanuchus cupido)  Ec
 lesser prairie-chicken (Tympanuchus pallidicinctus)  Not listed
 short-tailed albatross (Phoebastria albatrus)a  E
 Hawaiian petrel (Pterodroma sandwichensis)a  E
 pink-footed shearwater (Puffinus creatopus)a  Not listed
 buller's shearwater (Puffinus bulleri)a  Not listed
 Hawaiian coot (Fulica alai)  E
 bristle-thighed curlew (Numenius tahitiensis)a  Not listed
 red-legged kittiwake (Rissa brevirostris)a  Not listed
 Xantus's murrelet (Synthliboramphus hypoleucus)a  Not listed
 red-cockaded woodpecker (Picoides borealis)  E
 black-capped vireo (Vireo atricapilla)a  E
 elepaio (Chasiempis sandwichensis)  E
 Florida scrub-jay (Aphelocoma coerulescens)  T
 pinyon jay (Gymnorhinus cyanocephalus)  Not listed
 bendire's thrasher (Toxostoma bendirei)a  Not listed
 omao (Myadestes obscurus)  Not listed
 bicknell's thrush (Catharus bicknelli)a  Not listed
 sprague's pipit (Anthus spragueii)a  Not listed
 Laysan finch (Telespiza cantans)  E
 Kauai amakihi (Hemignathus kauaiensis)  Not listed
 Oahu amakihi (Hemignathus flavus)  Not listed
 anianiau (Magumma parva)  Not listed
 iiwi (Vestiaria coccinea)  Not listed
 cerulean warbler (Dendroica cerulea)a  Not listed
 rusty blackbird (Euphagus carolinus)a  Not listed
 saltmarsh sparrow (Ammodramus caudacutus)  Not listed

To evaluate if patterns found in previous NatureServe comparisons were evident in IUCN data (IUCN 2009), we compared classifications for all insects, crustaceans, gastropods, amphibians, and mammals evaluated by the IUCN in the United States. We studied classifications in animals because the IUCN has evaluated many more animals than plants or fungi, and we selected the six animal groups because they represent a broad sample of taxonomy, distribution, and habitats. The IUCN has not yet evaluated all U.S. resident insects, crustaceans, or gastropods, so our comparisons for these groups are not as representative as for birds, mammals, or amphibians. Nonetheless, the IUCN has evaluated more U.S. species of these groups than the ESA (IUCN 2009; USFWS 2009b), and our comparison gives baseline coverage of each group that should complement previous NatureServe comparisons.

Results

Birds

Of the 62 IUCN-listed birds in the United States, 25 species (1 Critically Endangered, 6 Endangered, 18 Vulnerable; 40.3% of the total) are not listed by the ESA (Table 1). Ten of the 25 species not listed by the ESA are endemic to the United States (40%). Species in IUCN categories of lower risk are more likely to be unrecognized: 5.3% of Critically Endangered, 42.9% of Endangered, and 62.1% of Vulnerable birds are not recognized by the ESA. Conversely, 23 bird species (29 total taxa including subspecies and populations) are ESA-listed as imperiled but not considered by the IUCN to be globally imperiled (six Near Threatened and 17 Least Concern; Table S1).

Twenty-three U.S.-resident bird species have gone extinct since 1825 (including one species, Corvus hawaiiensis, which survives only in captivity) (Figure 1). In addition, seven species are Possibly Extinct with the last confirmed sightings ranging from 1937 to 2004. Plotting the last confirmed sightings of Extinct, Extinct in the Wild, and Possibly Extinct birds by decade shows extinction peaks in the 1890s and 1980s (Figure S2). Of the 23 extinct species, 21 were endemic to Hawaii (as well as five of the seven Possibly Extinct species). Two species have been declared Extinct (Moho braccatus and Myadestes myadestinus), one Extinct in the Wild (C. hawaiiensis), and six Possibly Extinct (Numenius borealis, Myadestes lanaiensis, Psittirostra psittacea, Hemignathus lucidus, Paroreomyza maculata, and Melamprosops phaeosoma) since the passage of the ESA. Vermivora bachmanii was probably extinct when the ESA was passed, and the other species already had very small population sizes (with the possible exceptions of Myadestes myadestinus and Melamprosops phaeosoma).

Figure 1.

Hawaiian honeycreepers in peril. Extant species are in color; extinct and possibly extinct species are in grayscale. Five of the extant species shown (alauahio, Kauai amakihi, Oahu amakihi, anianiau, and iiwi) are IUCN-listed species that are unrecognized by the ESA. Numbers in parentheses specify how many species appear similar to the illustration. Note that akikiki is extant. Paintings and labels © H. Douglas Pratt, revised from Pratt (2005, Plate 7), used by permission.

Other animal groups

Our evaluation of the ESA's coverage of IUCN-listed insects, crustaceans, gastropods, amphibians, and mammals indicates that underrecognition of IUCN-listed species is not restricted to birds. We found 50% underrecognition for mammals, 80% underrecognition for amphibians, and 88.9–95.2% underrecognition for the invertebrates, which contributed to a mean of 74.1% underrecognition overall (Table 2). Vulnerable species (IUCN classification) were more often unrecognized (mean of 83.2%) compared to Critically Endangered (67.3%) or Endangered (64.9%) (Table 2).

Table 2.  Coverage of IUCN-listed animals (IUCN 2009) by the U.S. Endangered Species Act (USFWS 2009b). IUCN categories: CR =Critically Endangered, EN =Endangered, VU =Vulnerable. Percent of species that are unrecognized by the ESA are given in parentheses. For across-group totals, the mean percent of species unrecognized (± SE) is given
 Number of CR speciesCR species not recognizedNumber of EN speciesEN species not recognizedNumber of VU speciesVU species not recognizedNumber of species evaluated by IUCNTotal IUCN-listed speciesTotal unrecognized
Amphibians 22 (100)1713 (76.5)3629 (80.6)272 5544 (80)
Birds191 (5.3)146 (42.9)2918 (62.1)888 6225 (40.3)
Mammals 42 (50)207 (35)129 (75)451 3618 (50)
Gastropods6257 (91.9)3027 (90)10392 (89.3)458195176 (90.3)
Insects108 (80)1210 (83.3)8382 (98.8)207105100 (95.2)
Crustaceans1713 (76.5)3723 (62.2)135132 (97.8)203189168 (88.9)
Total114 83 (67.3 ± 14.2)13086 (64.9 ± 9.1)398362 (83.2 ± 5.8)2479 642531 (74.1 ± 0.09)

Discussion

Our data indicate that 40.3% of the U.S.'s IUCN-listed birds and more than 80% of lesser-known taxa have not been placed on the ESA list of endangered and threatened species. This underrecognition of species on one of the leading global lists suggests that the U.S. system is failing to keep pace with global listing assessments of imperiled species. It is unlikely that this classification gap can be attributed to species being stable in the United States but imperiled in their range outside the country. All unrecognized nonendemic birds (Table 1) have substantial proportions of their breeding and/or nonbreeding range in the United States. Possible exceptions are Pterodroma hasitata, Puffinus creatopus, and P. bulleri, but these three species are fairly common to common nonbreeding visitors to waters off the U.S. coast and therefore are eligible for listing even though they are not U.S. breeders. The ESA includes other nonbreeding species (e.g., Numenius borealis).

The ESA list includes 23 species of birds that are Near Threatened or Least Concern globally (Table S1). Nineteen of these species have only some populations or subspecies listed, which shows the ESA is protecting some regionally imperiled species. The remaining species, Somateria fischeri, Buteo solitarius, Charadrius melodus, and Dendroica kirtlandii, are ESA-listed in their entire range, but not by the IUCN, probably as a result of differences in listing criteria between the ESA and IUCN.

Bird species considered less-imperiled on the IUCN scale are more likely to not be listed under the ESA. Along these lines, Scott et al. (2006) found that nearly 80% of species listed by the ESA are endangered rather than threatened. There are several potential explanations for these patterns that are not mutually exclusive. The USFWS may: (1) list severely imperiled species first, due to an inability to consider all species at once, (2) primarily list species as a result of pressure from citizen petitions, which could focus on highly imperiled species, or (3) accept a higher risk of extinction compared to the IUCN. Risk prioritization seems to occur. Wilcove et al. (1993) found very small population sizes at the time of listing for 1,075 vertebrates and 999 invertebrates listed from 1985–1991, suggesting that species are not listed until they are highly imperiled. Outside pressure is also likely to be important. Petitions and/or lawsuits were involved with 71% of listings from 1974–2003 and have become even more important in recent years (Greenwald et al. 2006). In fact, the USFWS is so occupied with petitions and lawsuits from citizen groups that its ability to advance its own listing priorities is hampered (Stokstad 2005), and it requested a subcap to limit funding used to address petitions (USFWS 2011). Differences in risk tolerance may also contribute to classification differences between the IUCN and ESA. The ESA might be expected to list only highly imperiled species because listing results in legal protection, unlike the IUCN that has no legal enforcement ability in the United States.

This pattern of delaying listing until species are critically imperiled could be interpreted optimistically; at least the majority of species facing the greatest threat are protected. Unfortunately, chances of recovery are much reduced for highly imperiled species (Traill et al. 2010). The recent cases of two Hawaiian birds, akikiki Oreomystis bairdi and akekee Loxops caeruleirostris, are prime examples (Figure 1). Both were long known to be in serious trouble (listed by the IUCN as Endangered in 1994 and Critically Endangered in 2004 and 2008, respectively), but neither was listed by the ESA until 2010, while the akekee population continued to decline steeply (ABC 2008). Listing species before they reach critical imperilment would reduce extinctions and probably costs. It would be interesting for a future study to quantify the USFWS's savings from protecting species under the ESA when they are Vulnerable compared to Critically Endangered.

Our avian case studies (supporting information) exemplify USFWS decisions to not list declining, IUCN-listed species, and illustrate problems associated with vague categories, inadequate funding, and the warranted but precluded category. All three cases would have been more straightforward to resolve if clear, quantitative thresholds were included in the definitions of threatened and endangered. The effects of funding constraints were especially clear in the cerulean warbler's case where the USFWS took 6 years to reach a decision. The Kittlitz's murrelet case highlights the paradox of the warranted but precluded category; it seems unlikely that funds are so limited, or the Critically Endangered murrelet's priority is so low, that it should not be listed. While the USFWS is required to make a decision in 12 months, all three case study species experienced protracted listing times of 22 months to 6 years. These listing times are actually shorter than average; Greenwald et al. (2006) found the mean listing time for all species from 1974–2003 was >10 years.

In contrast to the avian case studies, the salmon case shows how the agencies can objectively and proactively list large groups of species by advancing their own listing priorities (supporting information). In the 1990s the NMFS coordinated teams of scientists to evaluate salmonids in Washington, Idaho, Oregon, and California. By 1999, the NMFS had listed 21 evolutionary significant units of salmonids as threatened and five as endangered. This case is an example of how science can be effectively translated to ESA policy. Public awareness of the value of salmonids for food and fishing likely contributed to the NMFS's comprehensive actions. Therefore, it seems reasonable that listing of other groups, such as unlisted birds in Table 1, could be accelerated if public interest in imperiled species increased (Schwartz 2008).

The multitaxa results suggest that underrecognition of IUCN-listed birds and mammals is less severe than in other, lesser-known groups (Table 2). This pattern could be explained if the USFWS accepts variable levels of extinction risk among taxa or if poorly known groups tend to be neglected (Wilcove & Master 2005). Wilcove & Master (2005) estimated that approximately 90% of the U.S.'s imperiled species (including animals, fungi, and plants) are not included on the ESA list. Given that Wilcove & Master's (2005) estimate was an extrapolation based on a few well-known groups, it is difficult to compare our results. Nonetheless, our finding of 74.1% underrecognition of IUCN-listed animals suggests the ESA covers more IUCN-listed species than NatureServe-listed species.

Our data indicate that a nearly 10-fold increase in listing would be required for the ESA to protect the gamut of IUCN-listed species. Considering the history and objectives of the two institutions, it is not surprising that the ESA covers fewer species. The Red List is intended to identify all imperiled species and has no regulatory apparatus. The ESA, however, legally protects species, so adding a species bears significant cost and responsibility to the agencies (funding per species is greater for the NMFS compared to the USFWS). The ESA is additionally influenced by politics because listing can have profound economic consequences (Ando 1999). If protecting all IUCN-listed species under the ESA is an unattainable endpoint, then triage could play a role in dictating listing decisions once all species are evaluated with objective and thorough procedures. A critical question under triage would be how to prioritize species based on endangerment, recovery likelihood, taxonomic uniqueness, and cost (Bottrill et al. 2008). We hold that listing a full complement of imperiled species under the ESA is not an insurmountable task.

Vague definitions of the threatened and endangered categories may also contribute to a lack of congruence between the ESA and IUCN lists (see “Introduction” for definitions). The ESA has been in place since 1973, but there is still ample room for debate on the meaning of these two key terms (Greenwald 2009; D’Elia & McCarthy 2010). There is a division between science and policy in ESA implementation by design, where science informs, but does not dictate, listing policy (Laband & Nieswiadomy 2006). In the case of the ashy storm-petrel, a lack of consensus when science informed policy delayed the listing decision and led to an outcome that is still contested by citizen groups and will likely incur further litigation costs to the USFWS. Such consequences from vague categories might be avoided if precedent quantitative thresholds were in place to guide decision-making when science is translated to policy. The IUCN uses unambiguous criteria, objective categories that measure probability of extinction, and a dynamic system that quantifies uncertainty in assessments (de Grammont & Cuarón 2006). Incorporating similarly quantitative attributes in the ESA decision-making framework would improve credibility of listing decisions and could reduce replication of effort between the USFWS and nongovernmental institutions such as the IUCN and NatureServe (Arroyo et al. 2009). Further, if ESA classifications eventually became more similar to IUCN methods, ESA data would be more useful for informing the Red List (Rodríguez 2008), which is an important function of national red lists to which the ESA does not currently contribute (Miller et al. 2007). Countries such as Singapore that use IUCN methods are able to evaluate hundreds of species in a few years (Davison et al. 2008); such rapid assessments could help reduce the backlog of ESA candidate species.

An increase to the ESA listing budget could speed the closing of the classification gap. External and internal observers agree that budgetary constraints are a primary barrier to listing species in a timely manner (GAO 1979; Stokstad 2005; Greenwald et al. 2006; USFWS 2006; Schwartz 2008). The protracted decision making in our avian case studies supports this conclusion.

Finally, we find that the warranted but precluded category compounds the classification gap by excluding imperiled species from the ESA. Warranted but precluded was created in 1982 to designate species that should be listed, but for which listing is currently precluded because of funding constraints (supporting information). While warranted but precluded findings can occasionally stimulate conservation efforts to prevent species from declining further (WGA 2011), this category has often been used by the USFWS as a loophole to slow listing (Greenwald et al. 2006). Given that citizen groups are unlikely to reduce pressure following warranted but precluded decisions, this category may be more likely to increase, rather than decrease long-term conservation costs.

In conclusion, our research agrees with previous findings that most of the United States’ imperiled species are not yet listed under the ESA. Our data indicate that less-imperiled (but at-risk) species are most likely to be overlooked, which does not bode well for the ESA's ability to mitigate declines before species become critically imperiled. Our avian case studies exemplify how a lack of consensus on key definitions, funding constraints, and the warranted but precluded category likely contribute to the classification gap between IUCN and ESA lists. By contrast, the salmonids case study shows how the agencies can proactively evaluate and list large groups of (albeit closely related) species.

Acknowledgments

M. Breed, P. Brewitt, S. Carvill, A. Chenault, Y. Ding Li, N. Greenwald, P. Levin, R. Medellín, J. Soberon, S. Wolf, and three anonymous reviewers provided helpful comments on the manuscript. We are grateful to D. Pratt for allowing us to reproduce Figure 1 and to P. Colla, R. Day, L. Hays, and D. Pereksta for photographs of the case study species. J. Griffiths and L. Collett assisted with the national red list and IUCN databases. JBCH and TCW were funded by EIPR scholarships.

Ancillary