FORECASTING FUTURE RESIDUAL VALUES The forecasting methodology for future values is based on an empirical approach using regression analysis with the principal variables for determining values being i) the age of the aircraft, (a proxy for utilisation) ii) economic and market conditions, and iii) the aircraft's position in the production cycle. We do not believe that including any other variable is of benefit. Problems begin to arise with maintaining independence, a prerequisite for regression analysis. Our basic future value figure is representative of the value which the aircraft should achieve with reference to the "normal" depreciation of the underlying asset as well as permanent impairments resulting from shifts in the particular types supply and demand curves, resulting from either political factors or airlines strategic decisions. We do not try to reflect cyclical factors at a particular point but rather we encapsulate the medium/long term trend in cyclical factors. In order to isolate the underlying trend our initial step is to eliminate the influence of inflation. Inflation serves to mask the diminution in the intrinsic value of the asset, and thus to observe the decline in the "real" value of the asset we need to step it out. After we removed the impact of inflation historically we have found the two most relevant cyclical factors in explaining aircraft values to be Real GNP and Real Oil Price year on year changes. Once we have removed the influence of these factors we can observe the trend and shifts of the underlying asset value and make inferences regarding the future. Thus our basic future model reflects the historical trend of the assets value as well as our qualitative judgements regarding the timing and magnitude of future permanent impairments. In order to turn this basic factor into an aircraft value we "add back" GNP and Oil price changes. We source our GNP and Oil price trends form the US Federal Aviation Administration (FAA) forecasts documents. A recognised and respected organisation whose annual forecast conferences are renowned for their content. In addition the aviation industry is a dollar denominated industry and approximately 50% of the worlds commercial aircraft are operated by US Carriers. Regression models of this nature are often termed "robust" i.e. they are not very sensitive to dramatic changes. It is also important to bear in mind that there is no guarantee that historical relationships will hold true in the future, however in our considered opinion they still offer the best long term forecasting methodology. Factors such as the number of aircraft in service, the order backlog, the concept of being a member of a family of aircraft, the quality and quantity of customers, all play a part in determining an aircraft's value. These factors are difficult to quantify in dollar terms, but we believe they are implicitly represented in the used sales price. Where possible we do try to quantify in monetary terms the maintenance and airworthiness directive / SB status of aircraft involved in transactions, since a strong relationship exists between the cost of conducting maintenance and value enhancement. For a type just entering commercial service, we believe the historical sales performance of aircraft from the same generic class which have had a wide acceptance and market, can give a reasonable indication as to the likely value retention of the new type, until such time that the subject aircraft has built up a statistically usable sales transaction base. Forecasting is not an exact science and the results from the statistical models (particularly those involving new types) are subject to critical qualitative review by personnel within Ascend, making use of the wealth of knowledge and experience. These checks particularly focus on the anticipated developments such as fleet roll-overs, and developments/changes in airline strategies, new aircraft pricing and new regulations over the period covered by the forecasts. In this respect the values projected are aligned for consistency with our internal aircraft delivery forecast and retirement / useful lives assumptions.   FORECAST SPECIFIC ASSUMPTIONS When taking Future Value forecasts, the following specific assumptions are made: The airframe manufacturers will continue to support their aircraft. The engine manufacturers will continue to support their engines effectively. The aircraft's design and construction is such that, given typical utilisation and reasonable care and attention, it will not incur undue maintenance and overhaul costs in comparison with other aircraft of similar class and age. All significant Service Bulletins (SBs) and Airworthiness Directives (ADs) are complied with. No new national noise legislation or other legislation will come into force which could adversely affect the aircraft and therefore their value. The aircraft under consideration are typical airline examples maintained in good condition and used in typical airline operations with an average annual utilisation and sector length as might be expected for this class of aircraft. The value determination is on the basis of the aircraft being in average 'half-life' condition as regards airframe and engine overhaul, equipment life etc. (Obviously, in reality, the aircraft will be continually moving around this hypothetical point.) It is also assumed that the aircraft will suffer no design or material defects or accident, the result of which would adversely affect its future value. Additionally, the value forecast is intended to reflect that which might be expected from the result of an 'arms-length single sale' transaction between a 'willing buyer and willing seller', free of any lease or charge. The possible result of forced sales has not been considered.   FORECASTING ACCURACY ACTUAL VERSUS FORECAST FUTURE VALUES Ascend has always exercised the best practices and used a high level of detail when forecasting. However, actual aircraft values inevitably will differ from the forecast values due to the following reasons Inherent Volatility in the Forecasting Model Desecration of the Base Scenario Assumptions, which can be sub-divided into Market conditions and Transaction conditions The aircraft differs from the specific assumptions set out above Each of these factors needs to be considered when producing a future value forecast. The Base Value as outlined above is the expected value however there is a wide distribution of potential values around this “Mean” value. MODEL UNCERTAINTY Ascend forecasts Base Values over various time horizons, and this forecast involves many factors as outlined in our Forecast Methodology. The actual Base Value may eventually differ from the forecast value due to four main reasons: Changes in Economic Assumptions Eventual changes in the timing and magnitude of step changes Non-adherence to the anticipated underlying depreciation trend Historical relationships failing to hold into the future To evaluate the impact of the errors, we constructed distributions comparing the forecasted base versus the actual values over three different forecast years. ECONOMIC CYCLE EFFECT Base values should not be biased by individual (Spot) factors of the economic cycle, they should however be affected by the long term trends in the cycle. Nevertheless the cyclic factors do impact spot market values, and hence the distinction between the base value and the spot or current market values occurs. As we can never be certain what stage the world will be in the economic cycle, this introduces volatility to the future values. Base values are determined assuming a balanced market, where Supply = demand We can quantify this aspect by looking at the historical relationship between market and base values for a particular group of aircraft and the graph below, illustrates that relationship. Using these two distributions Ascend has developed the AARQ model (Ascend Asset Risk Quantifier) analyse the potential impact of changes in the cycle and the deviation of the market from the base values. By taking a combination of these factors that impact an aircraft's value, a new distribution of potential aircraft values is produced. This is achieved using the mathematical Monte Carlo simulation technique, in effect to determine the likelihood of a future value being achieved.

VALUATION COMMENTARY

A fundamental element of the Ascend future value models is essentially based on the historical relationship between aircraft values, the age of the aircraft, fuel prices and the prevailing economic conditions, using historically the economic growth rates for the US economy as a proxy for world economic growth. This is particularly apt given the importance of the US carriers in the industry, they are and have been very much a barometer of the industry's well being. However, over the last 18 months we have achieved better results for some long haul types, most notably the 747-400, by using the Japanese economy as a proxy. Intuitively this is to be expected given the importance of the Asian market for this type. Value data points that can be useful to monitor development and structural shifts are not that numerous, indeed the growing lack of transparency should be of increasing concern to those closely involved with residual values. The US carriers were once legally bound to report transaction prices but this is no longer so. Confidentiality agreements are often invoked in deals concluded and with the rise in sale-leasebacks or deals with leases in place, this further clouds the issue. Historically it is also not unusual for aircraft to be in service for five years before any open market sale is noted. Indeed the 727-200Adv was not traded for 8 years after its introduction. With more 'structurally sound' aircraft, the 'age' factor in relation to values appears to be waning. We have witnessed a trend where the value differences between aircraft of different ages are narrowing. Historically some airframe manufacturers have gained a reputation for building aircraft with renowned structural integrity that eventually manifests itself in the behavioural pattern of aircraft values. Normally younger aircraft would carry a premium over old examples, albeit this slowly moderates over a useful life of perhaps 25 years. However, with aircraft such as the MD-80 this phenomenon is happening much quicker. The aircraft has a build quality that will ensure its economic useful life will end long before the structure wears out, hence there is no real reason to have a young airframe. Indeed with lease rates the market seem to polarise into pre and post 1990 builds. Thus the value differences due to age are smaller than comparable Boeing products, to the point where age is effectively immaterial.

ENGINE RESIDUAL VALUE FORECASTING

Engines are a "derived demand" product in that their greatest value derives from the ability to power an aircraft; therefore we adopt a different approach when valuing engines than when valuing aircraft. In our opinion, there are five factors that impact an engine's values: 1. The value of the Life Limited Parts: All of the engines under consideration have components installed with an explicitly defined life. Once the part has been utilised to its "Life Limit", it needs to be replaced. Intuitively any life remaining until the part needs to be replaced has some value. 2. Time remaining to Refurbishment: As an engine is utilised, its performance decreases. This results in increased fuel burn and possibly decreased reliability. Eventually the engine will need to be subjected to a shop visit to restore its performance. Therefore the closer an engine is to its optimal performance level, the greater its value. 3. Scrap / Carcass value: The metal that the engine is manufactured from and the electronics installed on the engine have an intrinsic value. 4. Guarantee: New engines carry guarantees from the manufacturer and these obviously add value. 5. Utility value: As well as the value of the physical parts an engine carries a value due to its ability to power an aircraft from point A to B and enables an airline to earn an economic return. Therefore, in theory and practice, all similar engines should have the same utility and scrap value if they are all capable of operating identically, additionally if the status of the life limited parts and the refurbishment is the same in terms of life consumed, their values should be identical. The only differentiating factor between a new and used engine, is the life consumed and the guarantee. However, not all engines consume life in a similar way, some are used as spare engines and may fly 200 hours a year, whereas others can be used for 4,500 hours per year. Therefore, if a generic engine is to be considered, it is much simpler and objective to assume that the engine is in a half-life status with regards to its LLP's and refurbishment This is the reason why we do not distinguish between years of build, but define a "used" engine. Additionally we do not wish to make subjective judgements about how a new engine is going to be utilised so we assume 'half-life' and adjust accordingly for guarantee, and first run. In addition, those 'technical factors' are also economic consideration in terms of price inflation, which are important contributors in the determination of an engine's value. The new list price of an engine and LLPs play a major part, for the former, it is while the engine remains in production, whilst for the latter, it remains an issue until the engine is broken for parts at which point demand for new LLPs tends to weaken. The real cost of an engine overhaul is an important variable in the value, particularly during Phase 1 and 2 and for the purposes of this report, we assume that the cost moves in lines with underlying inflation i.e. no real increase. A further point to note is that whereas we believe an aircraft depreciates more rapidly during the initial phase of its life and that fleet roll-overs negatively impact values, engines, conversely, tend to hold their value much better during their initial life phase, indeed values of overhauled engines closely correlate to the new list price of a spare engine. When first tier operators start to dispose of their fleets, this often provides a fillip to engine values. This occurs because first tier operators have large fleets and thus benefit from economics of scale in terms of the number of spare engines they require, but secondary operators will not have such large fleets and the economies of scale disappear in theory, creating a demand for more engines. This holds true until operators and owners start to retire their aircraft and thus supply increases once more and demand for spare engines and components declines rapidly. This can lead to a case whereby the only value an engine has, is in its maintenance condition.