Air Freight Economic Factors of Demand in Egypt

Introduction

Second World War caused a rapid increase of air freight transport supporting war efforts. More than 650000 tons of cargo carried by air between India and China from 1942 till 1945, and between 1948 and 1949 the largest air cargo in history lifted to support Berlin blockage. Since then air transport has been arranged by worldwide agreements which reinforced the international standards.

Development of air transport industry considers one of the main improvements in the twenty first century and one of the main factors of fast and reliable transportation mean. In the modern era, air transportation is a necessary mode of transportation due to its efficiency.

By 2012 air freight industry supported by 1738 freighters worldwide, 37% of them large freighter more than 80 tons air crafts, 36% medium size carry between 40 to 80 tons, and the standard size were 274% which carry capacity less than 45 tons (Kiboi et al., 2017).

In Egypt air transportation started in 1932 by establishment of Misr Airlines which carried in 1935 around 21830 kg of cargo flying 675067 KM. The study objective is identifying the main economic factors of air freight in Egypt during the period 1982 till 2019. Problem of study that air freight sector in Egypt needs further study as more than 99% of international freight depends on maritime transportation only.

The paper uses deductive approach deriving economic hypotheses. First hypothesis; exchange rate, and WPI "wholesale price index" as proxy of inflation rate both affect air freight negatively. Second hypothesis; exports as proxy of international trade and GDP affect air freight positively. The hypotheses examined empirically deploying Vector error correction model (VECM) for investigating economic factors of air freight through testing Further analysis carried using IRF "Impulse Response Function", and “Variance decomposition”.

The remaining of paper is covering brief discussion of air freight transportation in Egypt followed by main factors of air freight then theoretical and literature background, followed by presentation of data description employed in the model and explanation of methodology, followed by results of empirical analysis, then ending with conclusion.

Air Freight in Egypt

Air freight or transporting goods using air crafts mainly specialized in high value and light weight goods also perishable and short-lived goods, as fruits, and seafood which can't be preserved for long time (Feng & Yang, 2019).

Demand on Air freight recently increased due to the following main reasons:

First: development of microelectronics industry as medical equipment and products which raised air freight demand around 80 to 90%. High technology products accounts for a third of air freight and expected to increase in future.

Second: just in time philosophy which required products to reach markets quickly.

Third: increase in aircraft hull size and improvement of cargo handling systems and more efficient air cargo networks lead to reduction of costs and raised operational efficiency of air transport (Bozan, 2019).

In 1960s, air transport witnessed congestion in air traffic which induced companies looking for larger capacity airplanes. In 1970s, air industry developed massively by introducing large capacity air planes carrying larger volume with lower cost which increased demand on air transport (Kiobi et al, 2017).

Fourth: rapid growth of e-commerce cross-border transactions increased number of air parcels for example increased in China from 1.3 in 2010 to 6.9 trillion yuan in 2017 (Feng & Yang, 2019).

In Egypt, According to Boeing projects air freights fleet expected to increase by 60% from 2010 freighter in 2019 to 3260 in 2039, as air freight expected to grow by 4.1%. Air cargo traffic expected to more than double expanding from 264 billion in 2019 to reach 578 billion RTK (revenue ton-kilometer) in 2039.

Middle East share in air cargo increased from 4% in1999 to 13% in 2019 due to expanding their fleet of wide body passenger and freighter. International air traffic raised 9.8% between African countries from 2009 till 2019 with strong growth in cargo capacity in Egypt, represented 15% of international air freight in Africa in 2019 (Boeing, 2019).

According to IATA Egypt’s connections to Middle East was the fastest growing in the last five years. IATA 2019 report stated that Egypt’s air cargo facilitation through customs and borders regulations ranks 78th out of 124 countries according to ATFI "Air Trade Facilitation Index" and 43rd of 135 countries according to EFFI "e-Freight Friendliness Index".

The "Enabling Trade Index" ETI ranks Egypt 116th of 136 countries for free flow of goods across borders towards destination. The development of air freight sector in Egypt needs further study as more than 99% of international freight depends on maritime transportation only (OECD, 2020).

Main Factors of Air Freight Demand

The main factors affecting air freight demand in literature are GDP "gross domestic product", international trade and inflation. GDP as an indicator of economic growth has positive impact on air freight, trade increases with higher GDP levels which increase air freight traffic internationally.

Important factors affect air freight is inflation as higher level of prices of goods reduces purchasing power of money the reduce demand of goods and services. Worldwide data shows from 1990 to 2018 GDP tripled by rate of 279% and in same period demand of FTK "Freight Ton Km" as an indicator of air freight also tripled by rate of 239%. At same period import and export data increased by 4.5% and inflation rate decreased by approximately 70% (Alici & Akar, 2020).

That shows the importance of investigating the impact of price level on air freight demand using WPI and the impact of exchange rate as higher local currency value in comparison to other countries will make local goods and services more expensive relative to other countries which reduce exports and suppose to have negative impact on air freight. The study also is investigating impact of GDP and exports on air freight which supposed to affect air freight positively.

Literature Review

Based on Porter (2008), competitive advantage theory attractiveness and competitiveness of global economy identified upon main attribute. The factors of production that is divided into general factors consist of land, natural resources and unskilled labor, and the specialized factors consist of infrastructure, capital, and skilled labor. Those factors combination determines country’s competitive advantage in the global economic environment and influences freight needs, then air transportation in turn affects these attributes through a set of enabling mechanisms (Porter, 2011). This shows the importance of air transport to nations' economies, which require further study of the main factors of air freight demand in Egypt as an important source of country's competitive advantage.

To study main economic factors of air freight demand it's important to refer to neo-Keynesian Aggregate demand AD model which assumes negative relationship between AD and price level according to three effects. First is wealth effect as the purchasing power of money decline due to increased price level. Second, Net exports effect, in which higher price level, will reduce exports. Third is the interest rate effect, at higher price level, demand on loans increase to fulfill their requirement, which raise interest rate and decline AD (Parkin, 2014).

Empirical literature showed the importance of (GDP) and trade impact on Air freight as the study of Kasarda & Green (2005) studying the relation between air freight and both GDP per capita and trade in 63 countries. Also found bi- directional relationship between "GDP per capita and air freight". As well as positive significant relationship between industrial production and air freight as it increase exports. Yao (2005) using Granger causality and IRF "impulse response function" both confirmed a significant relationship between air freight and production of firms as well as input inventory and found mutual causality between "air freight demand and economic growth". Chang & Ying (2008) found strong relationship between GDP and air freight in Africa from 1970 to 2012. There is strong causality between air freight and trade as stated by Chiming (2008) states "GDP per capita, industrial added value, import and export" have significant impact on air transport.

Chang & Chang (2009) found bilateral relationship between "air freight and GDP growth" in Taiwan. Piecyk & McKinnon (2010) stated that demand of freight is mainly influenced by amount of produced and consumed goods. They argued that national economy expansion, increases overall demand of goods and services.

Aderamo (2010), studying air transport demand in Nigeria, results found that GDP, Inflation Rate and CPI are significant. Kupfer et al. (2011), studying Europe and Asia from 1983 to 2007, using pooled regression found strong positive relation between exports and air freight.

Yaru & Lina (2012) stated that economic development has a positive impact on air freight. Suryani et al (2012) studying Taiwan airport found strong effect on air freight demand. Yaru et al. (2012) state that economic development, domestic and international demand has statistical significant positive impact on air freight.

Chi & Baek (2013) studying air passenger and freight demand in USA found that economic growth has significant impact on both. Button & Yuan (2013) investigate causality between air freight and economic development in USA found that air freight stimulates economic development.

Yumeng (2015) found that foreign trade, GDP per capita and tourism have statistical significant impact on air transport. Hakim & Merkert (2016) found significance impact of GDP on air freight demand in short run while studying Asian countries from 1973 to 2014. Kiobi et al. (2017) studying economic determinants of air cargo in ten airlines investigated impact of "GDP, GDP per capita and interest rate". The study found positive significance impact of GDP growth rate and GDP per capita, which facilitate favorable business growth environment that raise exports and imports then raise demand of air freight transport.

Kupfer et al. (2017) including additional variables to model as cargo efficiency found that export strongly affect air freight demand. Zhang & Graham (2018) studying eight emerging countries from 1992 to 2014 found that GDP and exports are driving forces of air freight. In addition, literature confirms that travelling increases at high trade regions or countries (Kiraci and Battal, 2018).

Kiracı & Battal (2018), studying variables affecting air freight in Turkey from 1983 to 2015 using VAR analysis, creating 2 models for air passenger and air cargo which used explanatory variables as "foreign trade volume, industrial production index and foreign direct investment". The model found that GDP and industrial production index are statistically significant.

Cahyadin & Sarmidi (2019) found long-run co-integration between "export volume, labor force, external debt, and economic growth". Feng & Yang (2019) studied air transport development in China and found that airports number, GDP per capita, imports and exports have positive significant impacts on air passenger and air freight.

Bozan (2019) studying Far East and Turkey from 2008 till 2018 found that “GDP per capita” has influence on air freight demand. Alıcı and Akar (2020) studied air cargo in thirteen countries from 1980 to 2018 found that GDP affects air freight positively. İnan & Gökmen (2021) found a statistically significant relationship among air passenger and GDP.

Reviewing literature it's obvious that there are large scale of literature studying the relationship between air freight and GDP indicators and less literature studying export impact on air freight demand and very little papers studying impact of inflation. The current study will contribute to the existing literature by investigating impacts of inflation and exchange rate which also affect price level of exports. Also contribution to air freight literature in Egypt as there is no studies including air freight demand factors in Egypt which consider important as mostly international freight use maritime transport.

Data and Model Specification

The study will estimate "Vector error correction model" VECM investigating the economic factors of air freight in Egypt during the period from 1982 till 2019. Based on aggregate demand model price level should affect demand negatively the model use WPI "wholesale price index" as proxy of price level of goods, as assumed by economic theory rising inflation rate will affect exports negatively which will reduce air freight demand. GDP shows gross production level which supposed to affect the trade and air freight positively. Exchange rate expected to have negative impact on air freight as if local currency value is high relative to other countries local goods will consider more expensive than other countries which will reduce demand of exports that expected to reduce air freight demand. Trade supposed to have positive impacts as rising trade volume will raise international shipping modes which include air transportation the model use exports as proxy of trade.

All variables must be stationary of first level to use “Vector error correction model" VECM method; variables need to be checked for stationarity using unit-root test of “Augmented Dickey Fuller (ADF)”, and “Philips-Peron (PP)”. Then specifying the optimal lag length using “lag length criteria” and carrying diagnostics tests of VECM as “Breusch-Godfrey Serial Correlation Test”, residuals will be tested using “Breusch-Pagan-Godfrey Heteroskedasticity Test, and normality test”, and stability test using “inverse roots test” in which all inverse roots should be included inside the unit circle.

Then co-integration test used to examine co-integrated movement among variables in long run using Johansen and Juselius (1992) methodology, if variables found to be co-integrated further examination of dynamics relate variables can be carried using VECM methodology combining short and long run dynamics among studied variables. VECM can be written as follow.

∆y_t=Πy_(t-1)+∑_(i=1)^(p-1)▒Γ_i ∆y _(t-i) + ε_t (equ.1)

yt is "vector of included variables", ∆ is "difference operator", Γ is short run coefficients matrix, p is lag length. Matrix Π is the product of 2 matrices Π= αβ , β matrix of "stationary long run relationships" and α is "matrix of error correction terms". Second term of the equation presents short run dynamics as Γ is "coefficient vector of lagged first difference of variables" ∆ yt-i, i indicate VECM number of lags, ε is white noise error term.

As stated by Stock & Watson (1988) “If variables are co-integrated there exists at least one linear combination among them which is stationary, determined by coefficients of β matrix”. Time series used usually expressed in logarithm to be able to interpret β matrix as "long run elasticities". To identify causality direction ECT “error correction terms” is analyzed (α), which reveals speed of adjustment in which variables adjust to any deviations from long run equilibrium (Batool and Goodman, 2021).

Long run causality checked by usage of t-test for significance of adjustment speed in error correction term ECT.

Then short run causality is checked by deploying standard Wald statistic and granger causality test to determine causality direction. Granger causality as mentioned by Batool and Goldman (2021), “measures weather the current and past values of variable yt help to improve the forecast of future values of variable zt”.

Assessing variables responsiveness intensity to shocks in short and long run using "impulse response function" IRF analysis, “shocks denoted as one standard deviation in innovations the effect also transmitted to other endogenous variables through VECM dynamic structure, as IRF track effect of shocks on each innovation overall endogenous variables in the VAR system.

If innovation is simultaneously uncorrelated IRF can be directly interpreted so Cholesky decomposition is applied for making IRF innovations uncorrelated as they are usually correlated” (Mehmood et al., 2013). IRF traces the impact of each variable in the model to one shock on current and future values; it identifies responsiveness of dependent variable in VECM to a shock on error term. Further analysis of variables dynamics will be carried using "variance decomposition" VDC “which break down variance of unanticipated changes in dependent variable according to the contribution of each variable’s innovation” (Enimola, 2010). Applying VDC approach based on VECM for comparing influence magnitude among included variables in the model through the studied period.

Based on the above the current research will estimate VECM model for identification of short and long run dynamics that relate variables as shown in equation (2) air freight as target variable and explanatory variables as "exchange rate, WPI, exports, and GDP”.

∆ LNAIRFR_t = α0 + ∑_(i=1)^p▒α_1 LNAIRFR_(t-i)+ ∑_(i=1)^p▒α_2 ∆LNEXCH_(t-i) + ∑_(i=1)^p▒α_3 ∆〖LNEXP 〗_(t-i)+ ∑_(i=1)^p▒α_4 ∆〖LNGDP 〗_(t-i)+∑_(i=1)^p▒α_5 ∆WPI_(t-i) + φ 〖ECT〗_(t-1)+εt (Equ.2)

Variables defined as shown in table (1) in addition to descriptive data of each variable. ECT "refers to error correction term relates to the fact that the deviation from long run equilibrium is corrected gradually through series of partial short run adjustments", its coefficient (φi) is the speed of adjustment it measures speed at which dependent variables bounce back to equilibrium after a change in independent variable. εt is white noise error terms. Short run dynamics captured by coefficients (αi) of explanatory variables.

Table 1: Variables Abbreviation and measurement indicators

variable Indicator Max. Min. Std. Dev. Jarque-

Bera Prob. Obs.

LNAIRFR “Air transport freight -million ton-km” 6.181 3.980 0.559 1.821 0.402 38

LNEXCH “Official exchange rate” 2.878 -0.357 0.955 1.943 0.378 38

LNEXP "Exports of goods and services" 25.226 22.661 0.782 3.019 0.221 38

LNGDP “Gross Domestic Product constant” 26.710 25.030 0.495 2.382 0.304 38

WPI “Wholesale price index” 126.777 6.488 39.975 3.470 0.176 38

Source: Author's Estimation "Ln" stands for logarithm

Model Estimation

To avoid "spurious correlation" time series has to be stationary of same order or integrated of same order, in case of integrated of same order it can be modelled by VECM. The most commonly used stationary test are “Augmented Dickey-Fuller (ADF)” and “Philips-Peron (PP)” which are employed as shown in table (2), all examined series are I (1) integrated of first order.

Table 2: Unit Root Tests for stationarity

Variable PP ADF Integration order

 Level Differenced Level Differenced

LNAIRFR -2.1657

(0.2216) -11.7495

(0.0000) -2.0251

(0.2753) -7.0643

( 0.0000) I(1)

LNEXCH -0.6014

( 0.8583) -2.999

( 0.0445) -0.6014

(0.8583) -3.4524

(0.0154) I(1)

LNGDP -1.2254

(0.6531) -3.6706

(0.0089) -0.7157

( 0.8291) -3.2259

(0.0273) I(1)

WPI -0.0733

(0.9445) -4.0726

( 0.0034) 0.0432

( 0.9562) -3.9845

(0.0042) I(1)

LNEXP

 0.1095

(0.9623) -5.0307

(0.0002) 0.1794

(0.9675) -5.0721

(0.0002) I(1)

Source: Author's Estimation

The appropriate lag length, using optimal lag length criteria is the second lag as shown in table (3), which confirmed by diagnostics test that carried based on VECM as “Breusch-Godfrey Serial Correlation Test, Breusch-Pagan-Godfrey Heteroscedasticity Test, and normality test” which shows that the model is free of Heteroscedasticity and autocorrelation problems and normally distributed at second lag. Also the model found to be stable using inverse roots test as shown in figure (1).

Co-integration test examined co-movement in long run of studied variables using Johansen and Juselius (1992) methodology indicated as shown in table (4) evidence of long run relationship among the model's variables, based on second lag as found by "trace test and Max-eigenvalue" which found 2 co-integrated equations for simplicity the model is specified with only one co-integrated equation, imposing restriction of one long run relationship.

Table (3) Lag length Criteria and Diagnostics tests results

Lag Lag length Criteria Serial Correlation LM Tests Jarque-Bera

 Heterosk.

Chi Sq.

 LogL LR FPE AIC SC HQ LRE p- value p- value p- value

1 69.926 342.740 0.000 -2.281 -0.948* -1.821 18.798 0.828 0.000 0.025

2 101.959 43.931* 0.000 -2.683 -0.239 -1.840 25.007 0.501 0.423

 0.163

Source: Author's Estimation

Figure (1): Inverse Roots

Source: Author's Estimation

Table (4) Co-integration Rank Test

Hypothesized Trace Max-Eigen

No. of CE(s) Eigen-value Stat. Critical-Value Prob.** Statistic Critical- Value Prob.**

None * 0.774 100.715 69.818 0 52.153 33.876 0.0001

At most 1 * 0.550 48.561 47.856 0.0428 27.988 27.584 0.0444

At most 2 0.321 20.572 29.797 0.3848 13.585 21.131 0.4000

At most 3 0.152 6.987 15.494 0.5791 5.779 14.264 0.6417

At most 4 0.033 1.208 3.8414 0.2716 1.208 3.841 0.2716

Trace test and Max-eigenvalue test indicates two co-integrating equations at 0.05 level

Source: Author's Estimation

VECM employed for further identification of the dynamics that relate variables. VECM estimated for short and long run dynamics with air freight as target variable and explanatory variables as "exchange rate, WPI, exports, and GDP”. Shown in equation (4) and ECT shown in equation (5):

ΔL〖NAIRFR〗_t = -0.44 〖ECT〗_(t-1) + 0.40 〖ΔLNAIRFR〗_(t-1)+ 0.07 〖ΔLNAIRFR〗_(t-2) -0.51 ΔLNEXCH_(t-1) + 0.03 ΔLNEXCH_(t-2)+ 1.85 〖ΔLNEXP〗_(t-1)+0.88 〖ΔLNEXP〗_(t-2) -1.4 ΔLNGDP_(t-1)+ 8.15ΔLNGDP_(t-2) - 0.02 〖ΔWPI〗_(t-1) + 0.02 ΔWPI_(t-2)- 0.31 (Equ. 3)

〖ECT〗_(t-1) = 〖LNAIRFR〗_(t-1)- 0.161 LNEXCH_(t-1) + 1.76〖 LNEXP〗_(t-1) - 1.72〖 LNGDP〗_(t-1) – 0.028 WPI_(t-1)+ 10.996 (Equ. 4)

R squared (coefficient of determination) shows that almost 66 percent of total variations in air freight explained by explanatory variables. P-value of F-statistics is significant at 1% which shows that data is fitted good, and Durbin-Watson statistic is almost 2 which shows that the model is free of serial correlation, which also confirmed by diagnostics tests in table (3).

ECT in table (5) indicates long run relationship among variables as the coefficient is with correct sign (-0.44) and statistically significant at 5% confidence level, any deviation from long run equilibrium will be corrected at rate 44%.

Short run dynamics captured by coefficients of explanatory variables as shown in table (5) coefficients of independent variables shows that exchange rate is negative statistically significant at 5% confidence level to air freight as expected from literature as reduction of exchange rate of local currency will make local prices of goods and services relatively cheaper than other countries which will raise exports that induce increase in air freight as international transportation mean.

Exports found to be positive significant at 5% confidence level which goes with literature as higher level of exports will raise demand on airfreight. That goes with empirical literature Kiobi et al. (2017) study found positive significance impact of GDP growth rate, which facilitate favorable business growth environment that raise exports and imports then raise demand of air freight transport which also goes with the model estimates which found that GDP is positive statically significant at 1% confidence level to air freight which goes with economic theory.

Increased gross production will raise international trade and then international transportation, also goes also with empirical literature as found also by Kiracı & Battal (2018). WPI as proxy of inflation found to be negative statistically significant at 1% confidence level to air freight which goes with economic theory that there is negative relationship between prices and demand.

These findings confirmed also through using Wald Statistics and granger causality test as shown in tables (6 and 7) which both found that in short run there is unidirectional causality from Exchange rate, Exports, GDP and WPI to air freight at 5% confidence level.

Table (5) VECM Estimation

 Coef. Std. Err t-Stat. Prob.

ECT -0.44012 0.187915 -2.34215 0.0291

D LNAIRFR(-1) 0.400507 0.265995 1.505691 0.1458

D LNAIRFR(-2) 0.074644 0.16648 0.448364 0.6581

D LNEXCH(-1) -0.51237 0.198791 -2.57743 0.0176

D LNEXCH(-2) 0.037982 0.22031 0.172405 0.8648

D LNEXP(-1) 1.850334 0.538297 3.437388 0.0025

D LNEXP(-2) 0.880521 0.416433 2.114437 0.0466

D LNGDP(-1) -1.43258 2.191155 -0.6538 0.5203

D LNGDP(-2) 8.152072 2.452919 3.323417 0.0032

D WPI(-1) -0.02589 0.006672 -3.88123 0.0009

D WPI(-2) 0.029036 0.008009 3.625377 0.0016

C -0.31586 0.174684 -1.80815 0.0849

R-squared 0.660976 Durbin-Watson stat 1.9295

F-stat. 2.146627 Prob. (F-stat.) 0.059

     Source: Author estimation. D is first difference

Table (6) Wald Statistic Results

Wald statistic results

Dependent variable: D(LNAIRFR)

Variable F-stat. Chi-square

 value Prob. value Prob

LNEXCH 3.737 0.0409 7.473 0.0238

LNEXP 6.288 0.0072 12.576 0.0019

LNGDP 5.533 0.0117 11.066 0.004

WPI 9.842 0.001 19.684 0.0001

                            Source: Author’s estimation

Table (7) VEC Granger Causality

 Dependent variable: D(LNAIRFR)

Excluded Chi-sq df Prob.

D(LNEXCH) 7.473 2 0.0238

D(LNEXP) 12.576 2 0.0019

D(LNGDP) 11.066 2 0.004

D(WPI) 19.684 2 0.0001

Source: Author’s estimation

Impulse Response

The results are checked further with IRF “Impulse-Response functions” based on VECM. IRFs identify responsiveness of dependent variables in the VECM due to a shock on error term “it traces effect of one-time shock to one of innovations on current and future values of dependent variables". The results of IRFs for 37 years on yearly basis in Figure (2) showing a one standard deviation shock to exchange rate and WPI each causes significant decreases in air freight over the 37 studied years, while one standard deviation shock to GDP and exports each causes significant increases in air freight over the 37 studied years.

Variance Decomposition

To compare extent of contributions of time series FEVD “Forecasting error Variance decomposition” based on VECM gives "the percentage of unexpected variation in each variable that is produced by shocks from other variables comparing magnitude influence among studied variables". FEVD results at table (8) shows that in first year 100% of forecast error variance in air freight is explained by itself further we move in future over long run influence decreased to 63% while influence of other variables become stronger in predicting air freight, contribution of exchange rate on predicting air freight increased to reach almost 12%, exports contribution increased to reach almost 11%, followed by WPI contribution increased to reach almost 10% and the weakest contribution is GDP which increased to reach almost 4%.

Figure (2) Impulse Response Functions

 Source: Author estimation

Table (8): Variance Decomposition of LNAIRFR

Period S.E. LNAIRFR LNEXCH LNEXP LNGDP WPI

1 0.137224 100 0 0 0 0

2 0.229455 80.12542 4.549769 0.051787 0.291897 14.98113

3 0.261646 78.98188 8.527616 0.188387 0.662254 11.63987

4 0.307947 63.6826 9.646854 16.73211 0.518429 9.420005

5 0.329818 63.69739 8.674345 18.64096 0.630451 8.356851

6 0.364039 65.87675 7.987681 17.52739 0.534099 8.074077

7 0.40141 58.54127 7.759205 23.25006 2.031895 8.417567

8 0.419658 55.16805 8.026125 21.50385 4.120787 11.18118

9 0.431932 53.82725 8.511283 21.80497 5.036013 10.82049

10 0.449721 55.12163 9.643371 20.11892 4.648383 10.46769

11 0.467323 56.13068 10.13351 18.76426 4.401996 10.56955

12 0.480688 56.83812 10.29392 18.04956 4.312533 10.50586

13 0.491705 56.73854 10.35928 18.41383 4.408026 10.08032

14 0.506572 56.61838 10.16199 19.25243 4.294105 9.673098

15 0.5212 57.67914 9.976676 18.90287 4.130144 9.311173

16 0.535586 59.02025 10.04476 17.91373 3.941975 9.07928

17 0.54725 59.55765 10.18333 17.16911 3.93822 9.151686

18 0.557738 59.76321 10.22577 16.68914 4.025309 9.296578

19 0.567725 60.12491 10.34475 16.10968 4.095664 9.32499

20 0.579147 60.24334 10.53041 15.55056 4.16061 9.515083

21 0.589817 60.19042 10.71773 14.99719 4.314262 9.7804

22 0.599379 60.22869 10.94238 14.57377 4.408927 9.846239

23 0.608872 60.37423 11.15306 14.25341 4.419389 9.799909

24 0.619003 60.64258 11.25144 13.99993 4.374931 9.731118

25 0.629119 61.03243 11.29093 13.75651 4.315414 9.604725

26 0.638919 61.3985 11.32201 13.53776 4.266533 9.475193

27 0.648626 61.66431 11.31371 13.40694 4.238516 9.376527

28 0.658276 61.98267 11.29601 13.22892 4.204596 9.287794

29 0.667732 62.3568 11.32005 12.91414 4.171497 9.237511

30 0.677002 62.61634 11.36732 12.57622 4.169066 9.271053

31 0.685804 62.76307 11.42681 12.26822 4.203592 9.33831

32 0.694263 62.86595 11.5121 11.98183 4.246679 9.393438

33 0.702683 62.94647 11.60867 11.708 4.285765 9.451104

34 0.711088 63.03457 11.69863 11.4579 4.314058 9.494846

35 0.719374 63.14881 11.78464 11.24304 4.325662 9.497861

36 0.727611 63.27507 11.85428 11.07461 4.322968 9.473069

37 0.73591 63.41792 11.89207 10.95326 4.308394 9.428358

Source: Author estimation

Conclusion

Air freight consider important attribute of county's competitive advantage based on Porter (2008), which shows the importance of studying air freight specially in Egypt as 99% of international freight in Egypt relies only on maritime transport. That reduces the economic impact of Air freight that stressed on the importance of studying the main economic factors of air freight demand in Egypt.

Based on neo-Keynesian Aggregate demand AD model the study focused on investigating the impact of price level on demand of air freight using WPI as indicator of inflation. Impact of exchange rate as higher local currency value in comparison to other countries will make local goods and services more expensive relative to other countries which reduce exports.

In addition to studying GDP and exports impact which supposed to have positive impact on air freight demand due to increase domestic production which increase exports.

Using VECM approach found long run relationship between the studied variables, any deviation from long run equilibrium will be corrected at rate of 44%.

Short run dynamics captured by coefficients of explanatory variables shows that exchange rate is negative statistically significant as expected from literature as higher local currency value will raise local goods prices which reduce the demand on exports followed by air freight demand.

Exports found to be positive significant at which goes with literature as higher level of exports will raise demand on airfreight. That goes with empirical literature Kiobi et al. (2017) study found positive significance impact of GDP growth rate, which facilitate favorable business growth environment that raise exports and imports. That raise demand of air freight transport which also goes with the model estimates which found that GDP is positive statistically significant which goes with economic theory as increased gross production will raise international trade and demand on air freight.

WPI as proxy of inflation found to be negative statistically significant which goes with economic theory. These findings confirmed also through using Wald Statistics and granger causality showing unidirectional causality from Exchange rate, Exports, GDP and WPI to air freight.

The model further examined using IRF for 37 years on yearly basis a one standard deviation shock to each of exchange rate and WPI cause significant decreases in air freight, while shock to GDP and exports cause significant increases in air freight.

FEVD carried based on VECM which shows that 63% of air freight explained by its own innovative shocks followed by contribution of exchange rate of contribution rate of 12%, followed by exports contribution rate of 11%, then WPI with contribution rate of 10%, and the weakest contribution rate is GDP of 4%.

The findings of the study contribute to determination of factors affecting air freight demand, which is beneficial, to air freight industry and policy makers, contributing as well to the literature of air freight demand.

References

Aderamo, A. (2010). Demand for Air Transport in Nigeria. Journal of Economics July 2010 1(1).

Alıcı, A. and Akar, A. S. (2020). Macroeconomic Determinants of Air Cargo Demand: A Panel Data Analysis. Transport & Logistics. The International Journal, 2020; Volume 20, Issue 48, June 2020, ISSN 2406- 1069.

Batool, I. & Goldman, K. (2021). The Role of Public and Private Transport Infrastructure Capital in economic growth. Evidence from Pakistan. Research in Transportation economics 88(2021)

Boeing (2019). World Air Cargo Forecast 2020-2039

Bozan Z. (2019). Determinants of Demand for Air Cargo Transport between Turkey and The Selected Far East Countries. Ibn Haldun University, School of Graduate Studies, İstanbul.

Button, K., and Yuan, J. (2013). Airfreight Transport and Economic Development: An Examination of Causality. Urban Studies, Vol. 50, pp. 329–340.

Cahyadin, M., and Sarmidi, T. (2019). The Impact of Foreign Direct Investment, Labour Force, and External Debt on Economic Growth in Indonesia and Malaysia. Jurnal Ekonomi Malaysia, 53(1), 171-185.

Chang, C-P. and Ying, Y-H. (2008). The Generative Power of Air Freight in The Trade Openness – Economic Growth Nexus in African Countries. South African Journal of Economics Vol. 76:3. pp. 493-512.

Chang, Y-H. and Chang, Y-W. (2009). Air cargo expansion and economic growth: Finding the empirical link. Journal of Air Transport Management 15, pp. 264–265.

Chi, J. and Baek, J. (2013). Dynamic relationship between air transport demand and economic growth in the United States: A new look. Transport Policy, 29, 257-260

Chiming, G. (2008). The Evolution of Airport Distribution in China and Its Influencing Factors in the Past 50 Years-Based on Spatial Analysis and Mathematical Statistics. Economic Geography,2008(03): 445-449+474.

Enimola, S. (2010). Infrastructure and Economic Growth: The Nigeria Experience, 1980-2006. Journal of Infrastructure Development 2(2) 121-133.

Erraitab, E. (2016). An Econometric Analysis of Air Travel Demand: The Moroccan Case. European Scientific Journal, 12(7), 167-380.

Feng, M. and Yang, J. (2019). Analysis on the Influencing Factors of the Development of Air Transport Industry in China Advances in Social Science, Education and Humanities Research, volume 268.

Hakim, M. M., and Merkert, R. (2016). The causal relationship between air transport and economic growth: Empirical evidence from South Asia. Journal of Transport Geography (56), pp. 120–127.

IATA. (2016). Air Freight Market Analysis-June 2016. The International Air Transport Association (IATA).

İNAN, T., and GÖKMEN, N. (2021). The Determination of the Factors Affecting Air Transportation Passenger Numbers. International Journal of Aviation, Aeronautics, and Aerospace, 8(1).

Kasarda, J. D., and Green, J. D. (2005). Air Cargo as An Economic Development Engine: A Note on Opportunities and Constraints. Journal of Air Transport Management, 11(1), 459–462.

Kiboi, J., Katuse, P. and Zachary, M. (2017). Macroeconomic Determinants of Demand for Air Cargo Transport among Selected Airlines. Mosoti European Journal of Business and Strategic Management ISSN 2518-265X (Online) Vol.2, Issue 6 No.2, pp 20 - 37, 2017.

Kiracı, K. and Battal, Ü. (2018). Macroeconomic Determinants of Air Transportation: A VAR Analysis on Turkey. Gaziantep University Journal of Social Sciences, 17(4), pp. 1536-1557.

Kupfer, F., Meersman, H., Onghena, E. and Van de Voorde, E. (2011). Air Freight And Merchandise Trade: Towards A Disaggregated Analysis. Journal of Air Transport Studies, vol 2 (2), pp. 28-48.

Kupfer, F., Meersman, H., Onghena, E., and Van de Voorde, E., (2017). The Underlying Drivers and Future Development of Air Cargo. Journal of Air Transport Management, 61, pp. 6-14.

Mehmood, B., Shahid, A., and Younas, Z. (2013). Inter dependencies between Aviation Demand and Economic Growth in India: Cointegration Equation Estimation, Economic Affairs: 58(4):337-347.

Organization for Economic Co-operation and Development (2020). OECD Investment Policy Reviews: Egypt 2020, OECD Investment Policy Reviews, OECD Publishing, Paris

Parkin, M. (2014) Economics, Pearson Education Limited 11th edition

Piecyk, M. and Mckinnon, A. (2010). Forecasting the Carbon Footprint of Road Freight Transport in 2020. International Journal of Production Economics 128(1):31-42s

Porter, M. E. (2008). On competition. Harvard Business Press.

Porter, M. E. (2011). Competitive Advantage of Nations: Creating and Sustaining Superior Performance. Simon and Schuster.

Stock, H. and Watson, M. (1988). Variable Trends in Economic Time Series, Journal of Economic Perspectives vol. 2, no. 3, Summer 1988 (pp. 147-174).

Suryani, E., Chou, S-Y. and Chen, C-H. (2012). Dynamic Simulation Model of Air Cargo Demand Forecast and Terminal Capacity Planning. Simulation Modelling Practice and Theory 28, pp. 27-41

Valdes, V. (2015). Determinants of Air Travel Demand in Middle Income Countries. Journal of Air Transport Management, 42(1), 75-84.

Yao, V. W. (2005). The Causal Linkages between Freight and Economic Fluctuations. International Journal of Transport Economics, 32(2), 143-159.

Yaru, D. and Lina,P. (2012). Research on the Influencing Factors and Countermeasures of Air Freight Development. Air Freight Business, 2012(05): 31-34

Yaru, D., Lina, P., and Jianwei, S. (2012) Research on the Influencing Factors and Countermeasures of Air Freight Development. Air Freight Business, 2012(05): 31-34.

Yumeng, Q. (2015) Analysis of Development Factors of Air Transportation Based on Explanatory Structure Model. China Civil Aviation, (10): 106-107.

Zhang, X. and Graham, A. (2018), Patterns and Drivers of Demand for Air Transport. (Chapter 19) in book: The Routledge Companion to Air Transport Management (pp.313-330). Routledge.