U.S. Energy Information Administration - EIA - Independent Statistics and Analysis
Analysis & Projections
Outlook for Refinery Outages and Available Refinery Capacity in the First Half of 2014
Release date: February 27, 2014
With this report, EIA resumes regular semiannual reporting on refinery outages and their potential implications for available refinery capacity and petroleum product markets. EIA believes dissemination of such analyses can be beneficial to market participants who may otherwise be unable to access information regarding planned refinery outages.
This report uses the same analytical approach and the same data sources as in previous semiannual EIA reports on refinery outages beginning in 2008, up to and including Market Assessment of Refinery Outages issued in April 2011. Like prior reports, this report considers the adequacy of available refinery capacity to produce middle distillate (diesel, jet fuel, and heating oil) and gasoline to meet forecast demand for petroleum products, focusing on two refinery units, the crude distillation unit (CDU) and the fluid catalytic cracking unit (FCCU), that are most strongly correlated with distillate and gasoline production, respectively. The report estimates available refinery capacity using data on operable refinery capacity, planned refinery unit outages and historical average unplanned outages.
While continuity in approach and methods was appropriate for the restart of semiannual reporting on refinery outages, EIA recognizes that there have been significant changes in the structure of U.S. petroleum product markets and their relationship to global markets since 2008. U.S. refinery capacity and utilization rates have increased while U.S. demand for gasoline and distillate fuels, the major petroleum products, have generally declined. The United States, which was a national net importer of petroleum products throughout the interval when prior reports were produced, is now a significant net exporter of these products on a national basis, although the East Coast continues to remain a significant net importer of petroleum products.
Substantial changes in market conditions in recent years suggest that a report focused exclusively on measures of the adequacy of crude distillation unit (CDU) and fluid catalytic cracker unit (FCCU) capacity in relation to forecast demand for petroleum products, at the national and Petroleum Administration for Defense District (PADD) levels, may not provide sufficient insights for assessing petroleum product market conditions. For future reports, EIA plans to expand the analysis to include more detailed information about the adequacy of regional supplies of petroleum products, at both the PADD and sub-PADD level. EIA is also exploring expanding the analysis to include the impact of other refinery unit outages, e.g. hydrocrackers, on refinery production of gasoline and distillate, and to include a more detailed consideration of the impact of unexpected outages on product supplies.
Refinery outages result from the planned shutdown of refinery units for maintenance and upgrades and from unplanned shutdowns for a variety of causes, e.g. mechanical failure, fire and flooding. Planned maintenance is usually scheduled during times when refined petroleum product consumption is low, i.e., in the first quarter and again in the fall.
This report reviews the potential implications of refinery outages during the first half of 2014 on the supply of gasoline and middle distillate fuel oil (diesel, jet fuel, and heating oil). Estimates of potentially available refinery capacity (PARC) are developed by starting from total CDU and FCCU capacity levels nationally and in each PADD, then decrementing those capacity levels to take account of planned outages plus historical average unplanned outages (POPHAUO). Because actual unplanned outages realized during the first half of 2014 may differ, perhaps substantially, from historical averages, the actual level of realized average refinery capacity may differ significantly from PARC estimates.
One metric considered in this report compares estimated maximum input (EMI) capacity, which is directly proportional to PARC, to the level of capacity required to process U.S. estimated crude inputs needed to meet projected gasoline and middle distillate consumption, based on EIA's January 2014 Short-Term Energy Outlook (STEO) forecast1. Given the growth in refining capacity since 2008, coupled with the decline or stagnation in U.S. petroleum product demand since that time, it is not surprising that PARC-based EMI in each month through June 2014 exceeds the estimated monthly maximum inputs into CDUs required to produce the amount of products refined in the United States needed to meet U.S. petroleum product demand in the STEO forecast, as shown in Table 1. Table 2 shows that a similar result holds for FCCUs in the United States as a whole. Specifically, EMI calculated from PARC, after applying appropriate factors for the relationship between CDU and FCCU inputs and the historical maximum utilization of available FCCU capacity, indicates that aggregate U.S. FCCU PARC should be more than adequate in all months except May, when there is a small potential shortfall. The full report provides the U.S. total data presented in Tables 1 and 2 for each of the U.S. PADD regions. Adequacy of both CDU and FCCU capacity is more appropriately addressed on a regional rather than a national basis because aggregated data can mask regional capacity shortfalls.
Tables 3 and 4 summarize the surplus or shortfall of PARC-based CDU and FCCU EMI to meet estimated input requirements by month for each PADD. As shown in Table 3, CDU capacity should be more than adequate in all PADDs during January through February and May through June. Small capacity shortfalls may occur in PADD 1 in March and in PADDs 4 and 5 in April. However, barring higher-than-historical unplanned outages or consumption, there should be adequate supply from prior period surplus capacity/inventory and transfers from other regions to cover the shortfalls.
FCCU capacity, as shown in Table 4, should be adequate in PADDs 4 and 5 throughout the first half of 2014, barring higher-than-historical-average unplanned outages. Small capacity shortfalls may occur in PADD 2 in April and in PADD 3 in May and June, but there should be adequate supply from prior period surplus capacity/inventory and redirected trade flows to cover the shortfall. In PADD 1, there is a shortfall of FCCU capacity throughout the period. However, this shortfall does not take into account the volumes of gasoline that are regularly brought into PADD 1. PADD 1 received an average of 2.35 million bbl/d of gasoline from PADD 3 and imports combined during the first three quarters of 2013.
Actual petroleum market conditions in the first half of 2014 will depend on outage levels that are actually realized, both planned and unplanned, as well the availability of product supplies from inventories and the ability of markets to redirect and modify quantities of both planned imports and exports should they come under stress.
|Month||STEO-estimated CDU crude inputs1||Operable CDU capacity (stream day)2||Total CDU outages (POPHAUO in 2014)3||Available CDU capacity net of POPHAUO outages||Estimated maximum CDU crude inputs4||Surplus/(shortfall) (estimated maximum CDU crude inputs minus STEO-estimated CDU crude inputs)|
|Notes: 1January through June 2014 estimated CDU crude inputs are January STEO-forecast data. EIA's February STEO was published after the analysis contained in this report was conducted. Using February STEO data would not have a material impact on this report's findings. 2Operable CDU capacity (stream day) is the maximum number of barrels of input that a distillation facility can process within a 24-hour period when running at full capacity under optimal crude and product slate conditions with no allowance for downtime. Stream day capacity is typically ~6% higher than calendar day capacity. 3POPHAUO includes planned plus historical average unplanned outages. 4The estimated maximum CDU crude inputs are estimated by applying a factor (89%) to PARC that represents the highest 6-month rolling average utilization rate using data from January 2002 through September 2013, where utilization rates were estimated as crude inputs over stream day distillation capacity.
Source: January 2014 Short-Term Energy Outlook; Industrial Info Resources (IIR), February 1, 2014 database.
|Month||STEO-estimated CDU crude inputs 1||Â FCCU inputs based on CDU crude inputs2||Operable FCCU capacity||Total FCCU outages (POPHAUO in 2014)3||Available FCCU capacity net of POPHAUO outages||Estimated maximum FCCU inputs4||Surplus/(Shortfall) (estimated maximum FCCU inputs minus FCCU inputs based on CDU crude inputs)|
|Notes: 1January through June 2014 required crude inputs are STEO-forecast data. 2The FCCU inputs required are estimated by multiplying STEO-estimated CDU crude inputs by a factor (33%) that represents the average observed ratio between FCCU and crude inputs from January 2009 through September 2013. 3POPHAUO includes planned plus historical average unplanned outages. 4The estimated maximum FCCU inputs are estimated by applying a factor (91%) that represents the highest 6-month rolling average ratio between FCCU input volumes and capacity in the United States for the period from January 2002 through September 2013.
Sources: January 2014 Short-Term Energy Outlook; Industrial Info Resources (IIR), February 1, 2014 database.
|Month||PADD 1||PADD 2||PADD 3||PADD 4||PADD 5|
|Source: Industrial Info Resources (IIR), February 1, 2014 database.|
|Month||PADD 1||PADD 2||PADD 3||PADD 4||PADD 5|
|Source: Industrial Info Resources (IIR), February 1, 2014 database.|
As of early February, U.S. distillate (diesel fuel and heating oil) inventories are below the historical five-year average in all regions of the country except the West Coast. As a result, distillate inventories may be able to provide only a limited cushion should refinery outages during the first half 2014 exceed POPHAUO levels. The exceptionally cold weather that much of the United States experienced during January, which increased demand for space heating fuels, resulted in some unplanned refinery unit outages and reduced refinery runs, is one key driver of reduced distillate inventory levels. Lower inventories are also partly the result of changes in the distillate market. Increasing global demand for distillate fuel has changed the structure of the Ultra Low Sulfur Diesel (ULSD) futures curve, which has shifted incentives away from holding inventory for future consumption and toward exporting promptly into the growing global market for distillate fuel. EIA expects that growing transportation and industrial use will boost U.S. distillate consumption in 2014, but also expects that growth to be partially offset by a decline in heating oil use.
As of mid-February, U.S. gasoline inventories are at normal to high levels in most regions of the country. In PADDs 2 and 4, inventories are at the low end of the five-year range. EIA expects gasoline consumption will trend flat to lower in 2014, as the effects of an increasingly fuel-efficient light-duty vehicle fleet outweigh expected increases in consumption resulting from economic and population growth. With gasoline consumption projected to trend flat to lower and distillate consumption expected to be flat to higher, some surplus refining capacity is expected to be available to help meet unexpected demand changes during periods of refinery outages.
As a result of modest demand levels and adequate capacity after both planned outages and historical average levels of unplanned outages, and taking into account product supplies available from inventory as well as from redirected exports and imports, petroleum product supplies in the United States should be adequate at the level of refinery outages outlined in this report.
This report does not consider the impacts of refined product logistics and distribution, which could affect the movement of product to markets affected by unexpected outages.
1 EIA's February Short-Term Energy Outlook was published after the analysis contained in this report was conducted. Using February STEO data would not have a material impact on this report's findings.