ESA: Work Capability Assessments, Mandatory Reconsiderations and Appeals: June 2021
Published 10 June 2021
Applies to England, Scotland and Wales
The latest release of these statistics can be found in the collection of ESA: outcomes of Work Capability Assessments statistics.
This is a summary of the latest National Statistics on the outcomes of completed Employment and Support Allowance (ESA) Work Capability Assessments (WCA). The publication covers information on both initial and repeat ESA assessments in Great Britain.
This release includes:
- WCA outcomes of initial and repeat WCAs to end of December 2020
- WCA outcomes of completed Incapacity Benefit (IB) reassessments up to December 2020
- Completed initial WCA appeal outcomes for ESA claims with a start date to March 2020
- End to end initial ESA claim clearance times to end of December 2020
- Mandatory Reconsideration (MR) information to end of April 2021:
- (i) number of MR registrations
- (ii) number of MR decisions
- (iii) MR outcomes by type of decision
- (iv) Median MR clearance time
Figures for this release reflect the disruptions caused by the Coronavirus (COVID‑19) pandemic. It should also be noted that Universal Credit (UC) has now replaced income-related ESA for most new claimants and this is changing the volume and characteristics of ESA WCAs, MRs and appeals.
The last two releases consisted of a reduced set of tables due to issues causing delays with the receipt of the HM Courts and Tribunal Service (HMCTS) appeals data, which did not allow enough time for the inclusion of the appeals data. This issue has now been resolved and a complete historic timeseries is available.
1. Main stories
The statistics show:
- The number of completed ESA WCAs with a DWP decision in the latest quarter to December 2020 is 43,000.
- The number of completed initial ESA WCAs with a DWP decision in the latest quarter is 14,000, compared with 28,000 repeat ESA WCAs.
-
The majority, 28,000 (66%), of the total 43,000 ESA assessments completed in the latest quarter to December 2020 were repeat assessments
-
Outcomes for initial WCAs for the latest quarter to December 2020 show the majority (76%) resulted in a Support Group (SG) placement, followed by 14% to Work Related Activity Group (WRAG) and 10% Fit for Work (FfW)
-
The number of registered MRs has fallen from 310 in January 2021 to 240 in April 2021
-
The MR monthly median clearance time increased from 3 days in January 2021 to 7 days in April 2021
- The median end-to-end clearance time for claimants completing their initial WCA decreased, from 107 days in September 2020 to 96 days in December 2020
2. What you need to know
The sources of the statistics are:
- DWP’s benefit administration datasets including MR data
- Healthcare provider assessment data
- HM Courts and Tribunals Service (HMCTS) appeals data for completed appeals
Information is only available for claims that began at least 6 months following the assessment date, or 9 months following the date of claim. This is because of the time required for the customer journey to end and for the information recorded to be as complete as possible.
Universal Credit (UC) was rolled out to every Jobcentre across Great Britain by December 2018 and has now replaced income-related ESA for most new claimants. This has greatly changed the volume and composition of new claims to ESA.
COVID-19 operational easements
Figures for this release reflect the disruptions caused by the coronavirus (COVID‑19) pandemic, which has led to changes to operational procedures. Such changes were necessary to process claims and ensure people in need of financial help could access benefits, while keeping customers and staff safe.
DWP temporarily suspended face-to-face health assessments from 17 March 2020. This aimed at reducing the risk of exposure to coronavirus, hence safeguarding the health of individuals claiming health and disability-related benefits, many of whom are likely to be at greater risk due to their pre-existing health conditions. This means that people claiming ESA were not required to attend face-to-face assessments. However, paper-based assessments continued and telephone assessments were introduced from 7 May 2020. Since November 2020 video assessments started to be introduced. For more information of operational easements see the background information note .
The typical customer journey, prior to the COVID-19 easements, is detailed below:
- Claimant registers for ESA – payment issued at assessment rate
- Claimant referred for WCA and issued the “Capability to Work” questionnaire
- Claimant completes and returns questionnaire
- Assessment Provider (AP) may conduct face-to-face assessment or paper-based assessment and provides recommendation to DWP
- DWP makes a decision. Those found eligible for ESA are either placed in the WRAG, which offers support in preparing for work, or the SG if they are unable to work or complete work-related activity. Those not eligible are found FfW
- If the claimant disagrees with this decision, then depending on the nature of the decision they will either (i) have to request an MR (before they can appeal); or (ii) be able to appeal straightaway
Note: As a result of a High Court decision in July 2020, MR has been discontinued for any claimant who, having been found Fit for Work (FfW) following a WCA would, if they appealed, be paid ESA pending the outcome of that appeal. This is known as Payment Pending Appeal (PPA). Not all claimants qualify, and it applies only to Fit for Work disallowance decisions, not decisions such as disallowance due to Failure to Return a Questionnaire or Failure to Attend a WCA.
Rounding policy
Since March 2017, the following rounding policy has been applied to the statistical release. Please note that percentages shown within the release are calculated using figures prior to rounding.
From | To | Rounded to |
---|---|---|
0 | 1,000 | 10 |
1,001 | 10,000 | 100 |
10,001 | 100,000 | 1,000 |
100,001 | 1,000,000 | 10,000 |
1,000,001 | 10,000,000 | 100,000 |
10,000,001 | 100,000,000 | 1,000,000 |
3. Assessment volumes and outcomes: ESA initial and repeat assessments
Initial and repeat assessment volumes for all outcomes this quarter remain low, despite a 18% increase from the past quarter.
ESA and WCAs were introduced in 2008. Claimants who had previously been on Incapacity Benefit (IB), or who claimed ESA, would receive an initial assessment. If they remained on the benefit, they might then get repeat assessments to ensure they were still receiving the appropriate amount of support.
All completed assessments by quarter, December 2008 to December 2020
Source: Stat-Xplore: ESA Work Capability Assessments under ESA WCA by Completed Assessment.
In the latest quarter to December 2020, the majority (66%) of completed ESA WCA were repeat assessments. The remaining 34% were initial WCAs.
When ESA was introduced, the first WCAs were mostly initial assessments. At first, many of these related to claimants who were transferred from IB, and the winding down of the IB reassessment programme is reflected in the fall in the number of initial assessments after 2013. Repeat assessments are usually due between 3 months and 3 years after the previous assessment, and so the number of these increased gradually over the first years of the benefit, reaching a peak of 43,000 in May 2013. From January 2014, a period began in which repeat assessments were suppressed in order to focus on clearing initial assessments, hence managing the number of outstanding cases awaiting a decision and reduce delays for claimants. This reduced the overall number of assessments. During this period, claimants could still request a repeat assessment due to a change of circumstances, such as the development of a new condition or deterioration of the existing one.
From late 2013 onwards, the number of ESA WCA completed had risen substantially up to 270,000 in March 2017. This increase could be attributed to a variety of factors, including the focus on clearing initial assessments following changes in operational procedures and the re-introduction of repeat assessments in December 2015.
Since December 2018 when UC replaced most income-related new claims to ESA there has been a fall in the number of initial assessments required. This, together with the suspension of face-to-face assessments in March 2020 in response to the COVID-19 pandemic, has resulted in an overall downward trend in numbers of completed assessments. The quarter to December 2020 recorded 43,000 completed assessments, a slight increase (18%) when compared to the previous quarter. From September 2020 WRAG allocations could also be made via telephone assessment, which may start to have an effect on the number of completed assessments.
Initial and repeat assessment outcomes, December 2008 to December 2020
Source: Stat-Xplore: ESA Work Capability Assessments under ESA WCA by Completed Assessment.
Of the 14,000 initial assessments completed in the quarter to December 2020, 13,000 resulted in ESA entitlement.
In the quarter to December 2020 repeat assessments accounted for 66% of completed ESA WCAs, numbers increased by 15% from previous quarter to stand at 28,000.
The remaining 34% of completed assessments were initial assessments. Numbers of initial WCAs have been falling over the last couple of years, however the latest quarter to December 2020 saw a slight increase from 12,000 to 14,000. The overall decreasing trend in the number of initial assessments in recent years can be attributed to UC replacing most income-related new claims to ESA from December 2018 following the national rollout.
Assessment outcomes for ESA initial assessments by date of decision made, December 2008 to December 2020
Source: Stat-Xplore: ESA Work Capability Assessments under ESA WCA by Completed Assessment.
Outcomes of both ESA initial and repeat assessments for the latest quarter show the majority of ESA claimants are assigned to the Support Group (76% and 85% respectively).
Assessment outcomes for ESA initial assessments for the quarter ending in December 2020
ESA Initial | Percentages |
---|---|
Support Group | 76% |
Work Related Activity Group | 14% |
Fit For Work | 10% |
Assessment outcomes for ESA repeat assessments for the quarter ending in December 2020
ESA Repeat | Percentages |
---|---|
Support Group | 85% |
Work Related Activity Group | 14% |
Fit For Work | 0% |
Note that percentages may not sum to 100% due to rounding.
The latest trends for DWP decisions on WCA outcomes are impacted by the operational changes implemented in response to the COVID-19 pandemic, such as the suspension of face-to-face assessments. From March 2020 the focus was on initial assessments, however reassessments already present in the system were processed alongside those related to changes in circumstances, where possible. When telephone assessments were first introduced in May 2020 only SG decisions could be made, however this was extended to include WRAG decisions from September 2020.
Although the majority of ESA claimants for initial and repeat WCAs are allocated to SG for latest quarter, the proportions allocated to SG for initials have fallen from 84% last quarter to 76% in the quarter to December 2020, while proportions allocated to WRAG have increased from 4% to 14%. Similarly, for repeats SG allocations have fallen from 95% to 85% in the latest quarter, while WRAG have increased from 4% to 14%. This could be partly due the extension to allow WRAG decisions for telephone assessments from September 2020.
The proportions found FfW for both initials and repeat WCAs have remained low since last quarter. Falling from 12% to 10% for initial WCAs and 1% to 0% for repeats. It should be noted that the focus of WCAs was on initial WCAs and the proportions for repeats are based on very low numbers.
4. ESA WCA Mandatory Reconsideration registrations, clearances and clearance times
MR registrations have been following a downward trend, reaching 240 in April 2021.
By the end of April 2021, a total of 850,000 MRs have been registered. Of these, 99.6% had been cleared. The number of MR registrations and clearances within each month fluctuate over time:
- the number of MR registrations and clearances gradually increased between April 2013 and March 2017 as volumes of ESA customers increased
- MR registrations and clearances volumes peaked in March 2017 but have followed a downward trend since
- There were 240 MR registrations in the latest month, April 2021
When the MR process was first introduced, the number of MR decisions was lower than the number of MRs registered. As the MRs became established, clearance volumes started increasing and became comparable to registration volumes in recent years.
The number of ESA WCA MR registrations and clearances should be seen in context with the decrease of ESA claims as UC is rolled out, the changing composition of assessment outcomes, the effect of the COVID-19 pandemic and that, in some circumstances, MR has now been discontinued and claimants can appeal straightaway.
For the purpose of these statistics, we consider all reconsiderations occurring after the WCA are counted as MRs. For more information on what we have counted as an MR see section 10 of this release and the methodology note .
Mandatory reconsiderations registrations and clearances, October 2013 to April 2021
Source: Stat-Xplore: ESA Work Capability Assessments under Mandatory Reconsiderations – Registrations and Mandatory Reconsiderations - Clearances.
The median MR clearance time increased slightly to 7 days in April 2021.
The MR average clearance times is derived using calendar days by calculating the median of the time taken from the date the Benefit Centre (BC) has decided the MR to be valid, having considered any new information, until the date the decision is cleared by the decision maker at the Dispute Resolution Team (DRT). These dates are obtained from the DWP Decision Making and Appeals Case Recorder (DMACR) reporting system. The total clearance time therefore includes the time taken to transfer the case to DRT.
Immediately after the introduction of MRs, in October 2013, the median average time taken to clear MRs had sharply increased to stand at 37 calendar days in May 2014. MR clearance times have significantly decreased from their peak in May 2014, and have been quite volatile since, with peaks in September 2016 (15 days), August 2018 (16 days) and January 2020 (16 days).
For the month ending April 2021, it took on average 7 calendar days to clear MRs, an increase of 4 days since January 2021.
Mandatory reconsiderations median clearance times, October 2013 to April 2021
Source: Stat-Xplore: ESA Work Capability Assessments under Mandatory Reconsiderations – Clearances.
5. ESA WCA Mandatory Reconsiderations outcomes
Since April 2020 the number of MR decisions has remained low, with only 270 MR decisions made in April 2021.
The number of MR decisions had been following a decreasing trend since the beginning of 2019. With the changes brought in by the COVID-19 pandemic, the volumes of decisions fell sharply in April 2020. In more recent quarters, starting August 2020, there has been a slow and steady increase that peaked in March 2021 with 360 decisions made. The last month to April 2021 saw a slight decrease with 270 decisions recorded.
From October 2013 to October 2019, the majority of requests to reconsider ESA WCA outcomes were not revised at MR stage. Since then, the proportion of assessment outcomes where the award changed after the MR gradually increased, with some fluctuation. The proportion of MR decisions resulting in a change in award rose from 54% in February 2020 to 82% in April 2020, reflecting the new operational approaches as well as COVID-19 impact. Since May 2020, approximately two thirds of MR decisions led to a change in award. In the latest month, April 2021, 64% of decisions were revised.
Percentage of ESA WCA mandatory reconsiderations by outcome, October 2013 to April 2021
Source: Stat-Xplore: ESA Work Capability Assessments under Mandatory Reconsiderations - Clearances.
The increase in the proportion of revised decisions since 2019 should be seen in context with the large reduction in MRs, the new operational approaches to gather additional oral and written evidence at the MR stage, as well as the changing composition of WCA decisions influenced by a number of factors. These include the COVID-19 pandemic and the introduction of UC replacing most income-related new claims to ESA.
Mandatory Reconsiderations – Decisions by type for the quarter ending in April 2021
Decisions Made | Percentages |
---|---|
Customer failed to attend WCA, provide medical evidence or return questionnaire | 10% |
Customer disputes FFW decision | 6% |
Customer disputes ESA group allocations | 83% |
Others/Unknown | 1% |
Source: Stat-Xplore: ESA Work Capability Assessments under Mandatory Reconsiderations – Clearances.
Note that percentages may not sum to 100% due to rounding.
Mandatory Reconsiderations – Decision Outcomes for the quarter ending in April 2021
Decision Outcomes | Revised | Not Revised |
---|---|---|
Customer failed to attend WCA, provide medical evidence or return questionnaire | 66% | 34% |
Customer disputes FFW decision | 36% | 64% |
Customer disputes ESA group allocations | 68% | 32% |
Source: Stat-Xplore: ESA Work Capability Assessments under Mandatory Reconsiderations – Clearances.
The majority of ESA WCA MRs cleared in the quarter to April 2021 were made on disputes about ESA group allocations.
In the quarter ending April 2021 approximately 83% of MR decisions were made on disputes on ESA group allocation, followed by 10% on disputes where the claimant had not followed the claim procedures correctly. Of all MR decisions made this quarter only 6% were against FfW decisions.
Patterns for all the reasons why customers raise an MR are different in recent months, alongside a fall in the proportion of MR decisions for disputes regarding FfW outcomes. This could be attributed to the ongoing COVID-19 disruptions, as reassessment activity for some types of claims has not yet resumed and overall numbers of assessments remain low. Furthermore, as a result of High Court decision in July 2020, MR has been discontinued for any claimant who, having been found Fit for Work following a WCA would, if they appealed, be paid ESA pending the outcome of that appeal. Not all claimants qualify, and it applies only to Fit for Work disallowance decisions.
6. Appeals clearances and outcomes for initial ESA claims started up to March 2020
The number of appeals heard on initial FfW decisions continues to fall, reaching the lowest figure since the introduction of ESA in 2008.
Appeals outcomes on initial FfW decisions, December 2008 to March 2020
Source: Stat-Xplore: ESA Work Capability Assessments under Appeal Outcomes by Claim Start.
Being found FfW at initial assessment is the primary reason for claimants disputing a decision. These figures focus on FfW appeals for initial assessments.
The total number of appeals heard on FfW decisions for initial assessments is very low compared to pre-2013 figures, when mandatory reconsiderations were introduced. Figures have remained lower than pre-2013 levels over the last 4 reporting years.
There were very few appeal outcomes for claims that started in the latest three quarters to March 2020. The low numbers may be partly due to the decrease in FfW decisions, and an increase in MR revision rates since late 2019, which are likely to affect the number of claimants going on to appeal.
Note that these figures are grouped by claim start date, therefore numbers could increase as more appeals are completed for claims started in the most recent months.
7. Health conditions and ESA group allocation for initial assessments
‘Severe functional disability’ is the main reason for SG allocation this quarter.
Note that recent falls in volumes across all groups are likely to reflect falls in ESA initial assessments as UC rolls out.
There was a notable decrease in allocations to the ‘Physical or mental health’ risk group for claims started from October 2015 onwards. Volumes in this group fell from 28,000 in the quarter to September 2015 to 17,000 in the quarter to December 2015. This should be seen in context with the updated guidance on the application of risk introduced at the start of 2016.
For claims started in the quarter to September 2020, 1,500 cases were allocated to the SG due to ‘Physical or mental health risk’.
Most claimants assigned to the SG, who started their initial ESA assessment in the quarter to September 2020, were allocated due to health conditions linked to ‘Severe functional disability’. This accounts for 62% of all SG allocations this quarter.
Main reasons for SG allocations for initial assessments by claim start date, December 2008 to September 2020
Source: Stat-Xplore: ESA Work Capability Assessments under ESA WCA by Claim Start
‘Terminally ill’ numbers remained relatively low, accounting for just 1% of SG assignment reasons for claims started in the quarter to September 2020.
Note that, since 29 September 2017, claimants in the ESA SG no longer need to be reassessed if they meet the Severe Condition criteria. Information on this can be found on Stat-Xplore.
‘Adapting to change’ and ‘Social interaction’ remain the main reasons for WRAG allocation (with 15 points or more) this quarter.
Receiving 15 points or more is the main reason for assignment to the WRAG at initial assessment, however claimants can also be assigned to the WRAG at reconsideration or after appeal.
Main reasons for WRAG (15+ points) allocations for initial assessments by claim start date, December 2008 to September 2020
Source: Stat-Xplore: ESA Work Capability Assessments under ESA WCA by Claim Start.
For claimants allocated to WRAG with 15 points or more, the most common categories for scoring points this quarter were ‘Adapting to change’ and ‘Social interaction’. Of these claimants, 91% had an ‘Adapting to change’ condition and 88% scored points in the ‘Social interaction’ group.
Note that claimants can have multiple functional impairments, therefore appear in more than one category.
Other reasons for being assigned to WRAG (with 15 points or more) are less common (Upper Limb, Sensory, Continence, and Lower Limb).
8. Experimental data: Initial and repeat ESA WCAs for October 2013 to September 2020
Of the 4.8 million ESA claims with a start date between October 2013 and September 2020:
- 66% had a completed assessment; WCAs for the remaining claims are either still in progress or were closed by the claimant
- 400,000 MRs have been registered, following a completed WCA
- 99.8% of these MRs, which were raised after the WCA, have been completed, with the original decision revised 16% of the time
- 26% of claimants who raised an MR after the WCA went on to complete an appeal
- of the 100,000 appeals completed, 34% had the DWP decision upheld at hearing while the remaining 66% were ruled in favour of the claimant
Counts of cases going through the WCA, MR, appeals process for initial and repeat ESA WCA, October 2013 to September 2020
Source: Data tables: ESA outcomes of Work Capability Assessments experimental cohort statistics for initial and repeat ESA assessments, by period of claim start, October 2013 to September 2020.
As a result of High Court decision in July 2020, MR has been discontinued for any claimant who, having been found Fit for Work following a WCA would, if they appealed, be paid ESA pending the outcome of that appeal. Not all claimants qualify, and it applies only to Fit for Work disallowance decisions.
9. ESA WCA customer journey clearance times for initial claims (experimental statistics)
The median end-to-end clearance times for initial ESA WCA claims is 96 calendar days for the latest month in December 2020.
After 2008, the median end-to-end clearance time generally followed an upward trend, reaching a peak of 192 calendar days in August 2014. Since then, it has reduced considerably to 93 days in February 2016 and 79 days in December 2018. The end-to-end clearance time saw sharp increases in early 2019 (103 days in March 2019 and 107 in July 2019), to fall again to 70 days in September 2019. In the month to December 2020, the end-to-end median clearance times (claim registration to final DWP award decision) for initial claims decreased from 107 (in September 2020) to reach 96 calendar days. .
One of the stages of the end-to-end process accounts for the time between the customer’s referral to the Assessment Provider (AP) and the AP recommendation. This stage will usually involve a WCA assessment and includes the waiting time for the customer to complete and return the questionnaire. Multiple referrals are sometimes required before an assessment is completed and a recommendation received, as customers may not attend appointments or return questionnaires.
In December 2020, the median time between referral to the AP and their recommendation was 48 days. The AP clearance time has been fairly stable since September 2020.
Median clearance times for completed initial claims, October 2008 to December 2020
Source: Stat-Xplore: ESA Work Capability Assessmentsunder Clearance Times for Initial Claims.
Monthly clearance times (in weekdays) for individual stages of the ESA WCA process
Clearance Type | October 2020 | November 2020 | December 2020 |
---|---|---|---|
Claim Registration to WCA Referral | 24 | 26 | 23 |
WCA Referral to Assessment Provider (AP) Recommendation | 46 | 48 | 48 |
AP recommendation to DWP decision | 9 | 11 | 10 |
End-to-End ESA claim | 101 | 104 | 96 |
Note that the ‘end-to-end clearance times’ refer to time taken from claim registration to date of DWP decision. Claim registration date is the date the claimant has first contact with the department when making a claim to ESA. The DWP decision will include MRs where there is a completed decision.
10. About these statistics
The UK Statistics Authority has designated these statistics as National Statistics, in accordance with the Statistics and Registration Service Act 2007 and signifying compliance with the Code of Practice for Statistics. This badge does not currently apply to the experimental cohort figures or ESA clearance times.
Key uses of the statistics include:
-
providing the evidence base for assessing the potential effect of changes, monitoring and evaluation of DWP policy
-
answering Parliamentary Questions and Freedom of Information requests and forecasting benefit expenditure (in conjunction with expenditure statistics)
-
policy development and evaluation by local authorities and other welfare to work and pensions stakeholders and providers
This release only includes ESA WCAs.
This release does not capture:
- UC WCAs
- UC MRs
- UC appeals
Recent trends in these ESA WCA statistics will be affected by the roll out of UC.
Terminology
Term | Definition |
---|---|
Registration | Claimant registers an application for a WCA, MR or appeal. |
Clearance | DWP decision maker has determined whether the claimant should or should not be entitled to claim ESA. |
Mandatory Reconsideration | Claimant wishes to dispute a decision made on their claim and requests DWP to reconsider the decision. Due to operational practices performed within the DMACR system, some MRs are recorded on the source data as ‘reconsiderations’. Therefore, for the purpose of these statistics, all ‘reconsiderations’ occurring after the WCA (for reasons typically associated with MRs) have been included within the total number of MRs. |
MR clearance time | The clearance time begins from the point the MR is raised on the DWP administrative system by the Benefit Centre as a valid MR, having considered whether they can initially change the decision in the light of any new information. The total clearance time therefore includes the time taken to transfer the case to the Dispute Resolution Team and the time taken for the decision maker to make a decision. |
Repeat assessment | An existing claim that has been reassessed for ESA, as opposed to a new claim. A repeat assessment is the second or subsequent WCA undertaken on an existing, continuous ESA claim, usually between 3 and 24 months after the previous assessment. These claimants will have already been assessed as having a limited capability for work at their initial WCA and the repeat assessment will assess if their capability for work has changed. |
Future developments
UC WCA official statistics
DWP are currently developing official statistics on numbers of UC claimants who go through a WCA and related outcomes. When the information will be robust and accurate enough for publication as official statistics, we will preannounce its release in the GOV.UK release calendar.
IB reassessments information on Stat-Xplore
Information on Incapacity Benefit (IB) reassessments is currently published as excel data tables alongside the statistical summary on GOV.UK. With the September 2021 release, we are planning to publish IB reassessment data via Stat-Xplore alongside the data tables. For releases after September 2021, we will be releasing IB information only via Stat-Xplore.
Where to find out more
See Stat-Xplore for more detailed breakdowns of the data covering Region, Local Authority and Westminster Parliamentary Constituency breakdowns available for WCA outcomes by claim start date and completed assessment date.
Read the methodology note for more detailed information on these statistics.
Read more about ESA.
Contacts
DWP would like to hear your views on our statistical publications. If you use any of our statistics publications, we would be interested in hearing what you use them for and how well they meet your requirements.
For more information on these statistics, or to provide feedback on our publication, please email wca.statistics@dwp.gov.uk
For further enquiries, you can contact Press Office: 0203 267 5144
ISBN 978-1-78659-334-4