Quantifying the Effect of Non-Tariff Measures on Imports of Saudi Arabia Using a Panel ARDL Gravity Model

Abstract

Saudi Arabia implements a wide range of non-tariff measures on imports and exports. Different research articles have quantified the effect of non-tariff measures on trade, but their effect on Saudi Arabia has not been quantified. The major objective of this paper is to quantify the effect of non-tariff measures on the imports of Saudi Arabia. Panel data from 2000 to 2022 for four major regions trading with Saudi Arabia are used to estimate the panel ARDL gravity model. The results of the bound test confirm the presence of a long-run association between the model variables. In the long-run, the per capita income of Saudi Arabia is the main determinant of imports. In contrast, in the short-run the per capita income has no influence, and the non-tariff measures have a negative effect on import value. At the cross-sectional level, the results confirm the negative effect of non-tariff measures on the selected trade partners with varying degrees. The results ascertain the detrimental effect of the application of technical and non-technical measures on Saudi Arabia’s imports. We recommend policymakers in Saudi Arabia adopt a more transparent policy of NTMs application that leads to a sustainable supply of goods and services and ensures sustainable trade.

1. Introduction

Non-tariff measures (NTMs) are policy measures or policy interventions other than ordinary customs tariffs that affect international trade in goods at the border by changing quantities traded, prices, or both. They have discriminatory (protectionist) effects as they affect most of the traded goods and have more influence than trade tariffs [1]. NTMs include a wide range of instruments such as quotas, licenses, technical barriers to trade (TBTs), sanitary and phytosanitary (SPS) measures, export restrictions, custom surcharges, financial measures, and anti-dumping measures (

Table 1. International classification of non-tariff measures.

A	Sanitary and phytosanitary measures
B	Technical barriers to trade
C	Pre-shipment inspection and other formalities
D	Price control measures
E	Licenses, quotas, prohibitions, and other quantity control measures
F	Charges, taxes, and other para-tariff measures
G	Finance measures
H	Anti-competitive measures
I	Trade-related investment measures
J	Distribution restrictions
K	Restrictions on post-sales services
L	Subsidies (excluding export subsidies)
M	Government procurement restrictions
N	Intellectual property
O	Rules of origin
P	Export-related measures

Source: UNCTAD [2].

). The World Trade Organization (WTO) regulates NTMs under the agreements of the sanitary and phytosanitary (SPS) measures and technical barriers to trade (TBT), which allows member countries to adopt scientific measures to protect consumer, animal, and plant health or safeguard the environment. NTMs can be applied to imported goods only, or both imported and domestic goods. For example, quantitative restriction (QR) can be applied only to imported goods, while technical regulation applies to both imported and domestic goods.

NTMs represent a hidden cost of international trade that reduce world trade and make it less than expected [3]. The challenges associated with NTMs may include procedural obstacles such as long delays in testing or certification, inadequate facilities, lack of adequate information on regulations, or infrastructural challenges [4]. Many researchers investigated and confirmed the effect of unobserved trade cost on trade flows. For example, Anderson and Van Wincoop [5] showed that the costs associated with cross-border trade, even between well-integrated countries, were well above those that could be explained by geographic distance and traditional trade policies. Therefore, NTMs represent a trade policy challenge of ensuring that it is not restricting or distorting trade and is used for necessary and legitimate policy goals.

Sustainable trade occurs when the commercial exchanges of goods and services generate social, economic, and environmental benefits in accordance with the fundamental principles of sustainable development. International trade is recognized as an engine for inclusive economic growth, poverty reduction, and an important enabler to achieve sustainable development goals (SDGs). NTMs, mostly aiming to protect human, animal, and plant life, as well as the environment, can potentially, directly or indirectly, compromise the role of trade as a vehicle for sustainable development through less employment opportunities, less income, and less investment in production and infrastructure. As a result, societies become less prepared to pursue social and environmental dimensions of sustainable development [2]. For instance, Ngok [6] addressed the impact of NTM-SDG linkages on trade in the case of the fishery and aquaculture sector of the European Union, with results indicating that NTMs have predominantly trade-restricting effects, along with the positive impact of SDGs. A direct implication is that policies in this regard must be carefully designed with high levels of inter-governmental coordination.

Saudi Arabia is a member of G20 and is considered one of the emerging economies in the developing countries. Saudi Arabia’s economy largely depends on international trade; the ratio of merchandise and services trade (exports and imports) to GDP rose from 28% in 2000 to a peak of 88% in 2012 and reached 54% in 2022 [7]. The main factors affecting Saudi Arabia’s trade policy are its membership in the World Trade Organization (WTO) and the Gulf Cooperation Council (GCC) Treaty. As part of its WTO accession negotiations, Saudi Arabia bound 100% of its tariff lines, made extensive commitments under the GATT, became a signatory to the Information Technology Agreement (ITA), and notified sanitary and phytosanitary measures and technical barriers to trade adopted by public authorities.

Saudi Arabia applied a wide range of technical and non-technical measures to several products, mainly for health, security, moral, and religious reasons but also on SPS grounds. This paper focuses on the effect of NTMs applied by Saudi Arabia on trade by estimating how they affect the value of imported goods in Saudi Arabia. Using the ARDL approach, the gravity model is used to quantify the effect of NTMs on trade. The hypothesis to be tested is as follows: the application of NTMs has a negative impact on imports by impeding the entry of traded goods into local markets. The trade partners of Saudi Arabia are aggregated into four regions, the Gulf Cooperation Council (GCC), the European Union (EU), Asian countries, and North America (NA). The Saudi imports from the selected regions represent more than 90% of total imports, and, within GCC member countries, there is a free trade agreement.

2. Applied Non-Tariff Measures in Saudi Arabia

Three indicators—the frequency index, the coverage ratio, and the prevalence ratio—are used to identify the importance of trade measures and to assess their effects on Saudi’s trade flows. Frequency and coverage ratios provide a simple way of assessing the importance of NTMs in a country’s trade based on the inventories of NTMs [2]. The frequency ratio indicates how widespread NTMs are across different products, and the coverage ratio shows how much NTMs impact a country’s trade.

Importations from Saudi Arabia have a coverage ratio of 75.45% and a frequency ratio of 70.28% for non-tariff measures, compared with 71.9% and 43%, respectively, of all 75 covered countries. Exports of Saudi Arabia have a coverage ratio of 6.04% and a frequency ratio of 51.19% for non-tariff measures compared with 27% and 21%, respectively, of all 75 covered countries (

Table 2. Saudi non-tariff measures by sectors.

Sector	NTM Coverage Ratio	NTM Frequency Ratio	NTM Affected Product–Count	Import Values (USD Million)	Import Share (%)
Textiles and Clothing	100	100	729	5763	4.73
Animal	100	100	185	6529	5.36
Vegetable	100	99.68	309	8927	7.33
Miscellaneous	96.84	93.41	312	8124	6.67
Footwear	95.6	74.47	35	973	0.80
Transportation	94.11	74.19	92	25,331	20.80
Machines and Electric	90.29	87.39	672	38,8401	31.89
Food Products	84.88	94.62	176	6667	5.47
Stone and Glass	72.6	37.91	69	5167	4.24
Fuels	67.23	37.14	13	1105	0.91
Chemicals	60.5	33.7	247	8969	7.36
Plastic or Rubber	51.56	83.33	170	2901	2.38
Wood	46.92	46.93	107	1769	1.45
Hides and Skins	44.95	61.11	33	249	0.20
Minerals	5.05	25.26	24	77	0.06
Metals	2.32	32	176	415	0.34
All Import Products	75.45	70.28	3349	121,811

Source: https://wits.worldbank.org/ (accessed on 20 November 2024) [8].

). Animals, vegetables, textiles, and clothing are the most products exposed to NTMs as coverage and frequency ratios reached 100%.

Table 3. Saudi Arabia’s top ten most imposed non-tariff measures.

Sector	NTM Coverage Ratio	NTM Frequency Ratio	NTM Affected Product
Prohibition for TBT reasons (B110)	46.31	25.54	1217
Testing requirement (B820)	37.23	32.15	1532
Certification requirement (B830)	34.6	32.13	1531
Product quality or performance requirement (B700)	29.41	30.68	1462
Import license fee (F650)	22.39	20.13	959
Packaging requirements (B330)	19.68	31.69	1510
Marking requirements (B320)	18.81	29.95	1427
Product identity requirement (B600)	18.11	28.84	1374
Restricted use of certain substances in foods and feeds and their contact materials (A220)	17.64	14.44	688
Special authorization requirement for SPS reasons (A140)	17.21	18.03	859

Source: https://wits.worldbank.org/ (accessed on 20 November 2024) [8].

shows the top ten most imposed non-tariff measures in Saudi Arabia. Prohibition for TBT reasons, testing requirements, certification requirements, product quality or performance requirements, and import license fees are the most applied NTMs in Saudi Arabia.

3. Literature Review

Many studies investigated the effect of NTMs on international trade quantities and value. The motivation to focus on these measures is a mixed effect (positive or negative) proposed by the economic theory of how these measures affect the volume and direction of trade. For example, compliance with standards and regulations can raise producer costs but at the same time reduce consumer costs due to the availability of product quality information. Complying with NTMs requires product modification, testing, and special documents, which drive up cost and reduce profitability. The net effect of NTMs on trade (increase or decrease) depends on the net effects on supply and demand [9]. However, in the literature, the trade-impeding effect of NTMs is more common than positive ones [10]. One example of the negative effect of NTMs was when Saudi Arabia returned many shipments of sheep exported from Sudan due to incompliance with SPS measures, which cost the exporters more than USD 5 million.

Andriamananjara and others [11] studied the effects of eliminating certain significant categories of NTMs on the global economy using the CGE model. The results showed that the removal of the categories of NTMs under consideration results in global gains on the order of USD 90 billion. These gains occur particularly from the liberalization by Japan and the European Union by region and from the liberalization of apparel and machinery/equipment by sector (module 3). Khalid and Ghafor [12] investigated the protectionist effect of food safety standards on imports from selected developing countries, with a specific focus on certain products like food, tobacco, trunks, and machinery. They used the gravity model as an analytical framework. The results show that stricter regulations have a negative effect on trade, and the effect is even more significant if a developing country imposes standards like SPS and TBT regulations. Otsuki and others [13] quantified the impact of a new harmonized aflatoxin standard set by the EU on food exports from Africa using the gravity model. Their results suggest that the implementation of the new aflatoxin standard in the EU will have a negative impact on African exports of cereals, dried fruits, and nuts to Europe.

Shepotylo [14] studied the effect of non-tariff measures on extensive and intensive margins of exports in seafood trade by applying the gravity model. He found that the SPS measures increase the extensive margin of export and reduce the intensive margin, while TBTs reduce the extensive margin of export and increase the intensive margin. There is substantial heterogeneity of the response of exports to NTMs across HS six-digit product lines. Ghodsi and others [15] also studied the evolution of NTMs and their diverse effects on trade. The results show the impeding effects of NTMs on trade. There is a positive effect on the demand side compensating for the negative effect of increasing costs in the supply side, especially for SPS measures, while TBT measures are overall impeding trade. In a study of the welfare effect of non-tariff measures on trade, Beghin and others [16] proved that, when NTMs are reduced, welfare increases and trade expands, but the results showed that the reinforcement of a food safety standard can be socially preferable to the status-quo, both domestically and internationally. Bernini and others [17] investigated the impact of non-tariff barriers in Argentina, with a specific focus on non-automatic import licenses (NAILs) and their influence on firm exporting dynamics and employment of firms that rely on imported inputs. The findings showed that NAILs significantly reduce firm imports, leading firms reliant on these imported inputs to decrease exports and employment.

According to our knowledge and search, no available study quantifies the impact of NTMs on the trade of Saudi Arabia. This study is a contribution to fill this gap.

4. Methodology: Theoretical Framework

4.1. Data

The paper depends on comprehensive data on NTMs collected by the World Bank and other partners, where a good coverage of applied NTM data for Saudi Arabia and other countries are presented in witz.worldbank.org. The panel data of import values and quantities and GDP of Saudi trade partners are gathered from the Saudi General Authority for Statistics.

4.2. Gravity Model

Several methods can be used to quantify the effect of NTMs on trade such as the price gap method or an econometric-based method like the gravity model or CGE. In this paper we used the gravity model by applying the ARDL approach to quantify the effect of NTMs on Saudi total import value using panel data from 2000 to 2022. The gravity model is generally used to understand the bilateral trade among countries, and to predict the relationship between the size of the economy, the distance, and the value of trade, in addition to other trade measures like NTMs [18,19]. The gravity equation is considered consistent with a wide variety of theoretical foundations [20].

To test the hypothesis of the negative impact of NTMs on trade (total import value) in Saudi Arabia, the basic gravity equation includes measures that capture the effects of NTMs, distance, economic size, and free trade agreement as explanatory variables represented by:

$$\mathit{L}\mathit{n}{\mathit{Y}}_{\mathit{i}\mathit{j}}={\mathit{a}}_{\mathbf{0}}+{\mathit{b}}_{\mathbf{1}}\mathit{L}\mathit{n}{\mathit{G}\mathit{D}\mathit{P}}_{\mathit{i}\mathit{t}}+{\mathit{b}}_{\mathbf{2}}\mathit{L}\mathit{n}{\mathit{D}}_{\mathit{i}\mathit{j}}+{\mathit{b}}_{\mathbf{3}}{\mathit{N}\mathit{T}\mathit{M}}_{\mathit{i}}+{\mathit{e}}_{\mathit{i}\mathit{j}}$$

(1)

where Ln is the natural logarithm, Y_ij is the value of import of country j to Saudi Arabia (i), GDP_i is the per capita income of Saudi Arabia for period t, NTM is non-tariff measure included as a dummy variable starting from the application date (dummy = 0 before 2005, dummy = 1 for 2005 and onwards), where 2005 is the date of Saudi Arabia’s accession to the WTO. The challenge of analyzing the impact of NTMs is the lack of exogenous variation of their magnitude over a long period of time [6,21]; to overcome this problem, a dummy variable is used. The date of accession to the WTO is used because, within the WTO, member countries have to notify SPS and TBT adopted by public authorities. Disdier et al. [21] prove that SPS measures and TBTs have a negative impact on trade flows. This result is observed whatever the measure used for these NTMs, e.g., a simple dummy, a frequency index, or an ad valorem equivalent. D_ij is the geographical distance between two countries, which is a proxy for trade cost. To avoid using constant data for each section, distance is weighted by the value of import from each section. The e_ij represents the random error term. The a₀ term is a regression constant, and the b terms are coefficients to be estimated. In the above equation, parameter b3 measures the effect of NTMs on trade; positive or negative signs of the parameter reveal the trade-enhancing or trade-impeding effects of NTMs [22].

4.3. Empirical Model

The first step of estimating Equation (1) is to test for the stationarity of included variables using the panel unit root test. If the time series of import value and per capita income and distance are non-stationary at the level, and become stationary at the first difference, this means that the OLS method to estimate Equation (1) is not valid and can lead to spurious results. The suitable model to estimate Equation (1) is the panel autoregressive distributed lag (ARDL) approach, where cointegration and long-run association between model variables, especially import value and per capita income, are expected. The panel ARDL model is considered to be the best econometric method when the variables are stationary in I(0) or integrated in order I(1) [23]. One of the useful features of the ARDL model is its accounting for short-run and long-run effects, as well as for cointegration between variables [24].

Equation (1) can be rewritten in a panel ARDL form as follows [25]:

$$∆\mathit{L}\mathit{n}{\mathit{Y}}_{\mathit{i}\mathit{j}}={\mathit{a}}_{\mathbf{0}}+{\sum }_{\mathit{k}=\mathbf{1}}^{\mathit{n}}{\mathit{b}}_{\mathbf{1}}∆\mathit{L}\mathit{n}{\mathit{Y}}_{\mathit{k}-\mathbf{1}}+{\sum }_{\mathit{k}=\mathbf{1}}^{\mathit{n}}{\mathit{b}}_{\mathbf{2}}∆\mathit{L}\mathit{n}{\mathit{G}\mathit{D}\mathit{P}}_{\mathit{t}-\mathit{k}}+{\mathit{b}}_{\mathbf{3}}{\mathit{L}\mathit{n}\mathit{D}}_{\mathit{i}\mathit{j}}+{\mathit{b}}_{\mathbf{4}}{\mathit{N}\mathit{T}\mathit{M}}_{\mathit{i}}++{\mathit{\gamma }}_{\mathbf{1}}\mathit{L}\mathit{n}{\mathit{Y}}_{\mathit{t}-\mathbf{1}}+{\mathit{\gamma }}_{\mathbf{2}}\mathit{l}\mathit{n}{\mathit{G}\mathit{D}\mathit{P}}_{\mathit{t}-\mathbf{1}}+{\mathit{e}}_{\mathit{t}}$$

(2)

where Δ is the first difference and e_t is the white noise.

The Akaike information criterion (AIC) is used to choose the lag length. After applying the bound test and its finding prove the existence of the long-run association between variables, this study uses the error correction model (ECM) to find the short-run dynamics. The ECM general form of Equation (2) is represented by Equation (3) as follows:

$$∆\mathit{L}\mathit{n}{\mathit{Y}}_{\mathit{i}\mathit{j}}={\mathit{a}}_{\mathbf{0}}+{\sum }_{\mathit{k}=\mathbf{1}}^{\mathit{n}}{\mathit{b}}_{\mathbf{1}}∆\mathit{L}\mathit{n}{\mathit{Y}}_{\mathit{k}-\mathbf{1}}+{\sum }_{\mathit{k}=\mathbf{1}}^{\mathit{n}}{\mathit{b}}_{\mathbf{2}}∆\mathit{L}\mathit{n}{\mathit{G}\mathit{D}\mathit{P}}_{\mathit{t}-\mathit{k}}+{\mathit{b}}_{\mathbf{3}}\mathit{L}\mathit{n}{\mathit{D}}_{\mathit{i}\mathit{j}}+{\mathit{b}}_{\mathbf{4}}{\mathit{N}\mathit{T}\mathit{M}}_{\mathit{i}}++\mathit{\varnothing }{\mathit{E}\mathit{C}\mathit{M}}_{\mathit{t}-\mathbf{1}}$$

(3)

where ∅ is the coefficient of ECM for short-run dynamics. The ECM coefficient indicates the speed of adjustment in the long-run equilibrium after a shock in the short-run [23].

5. Empirical Results

Table 4. Statistical summary of the study variables.

	Lnimport (USD Million )	lnGDP (USD)	LnD (km)
Mean	29,260.90	19,178.52	5695.750
Median	23,073.75	20,037.80	4889.500
Maximum	222,317.9	34,454.13	12,110.00
Minimum	1077.299	8643.400	894.0000
Standard deviation	35,069.13	6757.746	4097.932
Skewness	3.257730	0.032300	0.553408
Kurtosis	15.45792	2.453346	2.019955
Jarque–Bera	757.6625	1.161514	8.377862
Probability	0.000000	0.559475	0.015162
Observations	92	92	92

The summary values of included variables are computed before conversion to logarithm.

shows the statistical summary of the variables used in this study. The average import value is USD 29.26 billion from the selected four regions, with a maximum value of USD 222.3 billion (import value from Asian countries) and a minimum value of USD 1.1 billion (import value from GCC). The per capita GDP average value is USD 19.1 thousand, and the maximum value of per capita GDP is USD 34.4 thousand, with a minimum value of USD 8.6 thousand. Saudi Arabia is a major oil exporting country with a high per capita income. The mean value of the geographical distance is 5695 km, with a maximum value of 12.1 thousand km for North American countries, and a minimum distance of 894 km for the GCC.

5.1. Unit Root Test

Before running the ARDL bound test, it is important to check for the unit root of included variables; all variables must be stationary at I(0), I(1), or both [23]. This study checked for the existence of a panel unit root by using the augmented Dickey–Fuller (ADF) and Phillips–Perron (PP) methods. The results of ADF and PP prove that the variables are non-stationary at the level and become stationary at the first difference or have I(1) order of integration (

Table 5. Panel unit root test of model variables.

Variable	Method	t-Statistic	Prob.	Cross-Section	Obs.	Order of Integration
	At level
Lnimport	ADF PP	10.488 7.009	0.2324 0.5358	4	84 88
lnGDP	ADF PP	0.875 0.526	0.998 0.999	4	84 88
LnD	ADF PP	10.488 7.007	0.232 0.535	4	84 88
	At first difference
Lnimport	ADF PP	48.616 71.263	0.000 0.000	4	80 84	I(1)
lnGDP	ADF PP	18.699 25.871	0.016 0.001	4	80 84	I(1)
lnD	ADF PP	48.616 71.263	0.000 0.000	4	80 84	I(1)

).

5.2. ARDL Bound Test for Cointegration

The bound test is used to prove the existence of cointegration between study variables, especially Lnimport and LnGDP, because NTM and LnD variables are treated as fixed regressors in the model.

Table 6. Bound test results.

Cross-Section	Obs.		F-Stat.		t-Stat.
1	22		47.19251		−9.250859
2	22		65.711		−11.29263
3	22		25.98869		−3.866673
4	22		23.92927		−6.469296
	10%		5%		1%
Sample Size	I(0)	I(1)	I(0)	I(1)	I(0)	I(1)
			F-statistic
30	−1	−1	−1	−1	−1	−1
Asymptotic	2.440	3.280	3.150	4.110	4.810	6.020

I(0) and I(1) are, respectively, the stationary and non-stationary bounds.

shows the results of the ARDL bound test. Calculated F-statistics must be larger than the upper and lower bound at 1% or 5%. The results confirm the existence of cointegration and a long-run association between Lnimport, the dependent variable, and LnGDP, as the calculated F-statistics for four sections are above the upper bound at 1%.

5.3. Long-Run and Short-Run Results of the Panel ARDL Model Using the Pooled Mean Group PMG Estimator

After verifying the existence of cointegration and long-run association between model variables, the pooled mean group (PMG) estimator is used to estimate panel ARDL and calculate the long-run and short-run effects. The PMG estimator constrains long-run coefficients to be identical but allows short-run coefficients and error variances to differ across groups [26]. A diagnostic test of the goodness of fit of the estimated model shows that error terms are normally distributed, and a higher value of log-likelihood indicates that the estimated model fits the data.

Table 7. ARDL estimates of long-run and short-run effects.

Variable	Coefficient	Std. Error	t-Statistic	Prob.
	Long-run (Pooled) Coefficients
Ln(GDP)	1.328972	0.014676	90.55661	0.0000
	Short-run (Mean-Group) Coefficients
$\varnothing$	−0.585099	0.132034	−4.431437	0.0000
NTM	−0.353038	0.030114	−11.72338	0.0000
LnD	−0.394331	0.079276	−4.974131	0.0000
Log-Likelihood	32.66145

presents long-run and short-run panel ARDL PMG estimation results. The long-run results show that LnGDP is the only variable that affects Lnimport, which means the per capita GDP (income) is the main factor that affects the import value of Saudi Arabia from included country groups in the long-run. The coefficient has the correct sign and is statistically significant. The long-run association indicates that, when per capita income increases by 1%, the import value increases by 1.3%.

In the short-run, the variables NTM and LnD affect the import value of Saudi Arabia. The coefficients of the variables have the correct signs and are statistically significant (Table 7). The application of the NTMs has a negative impact on import value in the short-run. The coefficient of NTM indicates that the applied NTMs are impeding trade and decreasing the import value of Saudi Arabia; the coefficient of NTM (−0.35) indicates that the import value of Saudi Arabia shrinks by 35% compared with the case where no NTMs are applied. The distance coefficient is negative and significant, as expected, which proves that the larger the distance between trading partners, the higher the cost of transportation.

The ECM coefficient (∅) shows the speed of adjustment in the cointegration equation; it should have a negative sign and should be statistically significant. In our model results, the estimated ∅ coefficient has the correct sign (negative) and is significant, so we can conclude that the included variables are cointegrated. The coefficient value of −0.58 means that about 0.58% of departure from long-run equilibrium is corrected each period.

5.4. Cross-Sectional Error Correction Model

Cross-sectional ECM is estimated after proving the existence of long-run associations between variables to show short-run dynamics across sections included in the model.

Table 8. Cross-sectional error correction model results.

GCC
Variable	Coefficient	Std. Error	t-Statistic	Prob.
$\varnothing$	−0.296815	0.072287	−4.106043	0.0006
NTM	−0.433437	0.143197	−3.026855	0.0069
LnD	−0.304038	0.087995	−3.455157	0.0027
2. Asia
$\varnothing$	−0.900082	0.076691	−11.73654	0.0000
NTM	−0.287392	0.053525	−5.369355	0.0000
LnD	−0.624585	0.057986	−10.77126	0.0000
3. EU
$\varnothing$	−0.456605	0.122540	−3.726179	0.0014
NTM	−0.345906	0.157734	−2.192976	0.0410
LnD	−0.277166	0.094420	−2.935465	0.0085
4. NA
$\varnothing$	−0.686895	0.099078	−6.932864	0.0000
NTM	−0.345417	0.101423	−3.405723	0.0030
LnD	−0.371535	0.058281	−6.374835	0.0000

shows the results of the cross-sectional ECM. The model results for four regions are consistent, as all included variables have correct signs and are statistically significant. The speed of adjustment, measured by the coefficient of OINTEQ (∅), has a relatively low value for GCC (0.29) and EU (0.45) regions and a high value for Asia (0.9) and North America (0.69). This means the GCC and EU need longer periods to attain the long-run equilibrium than Asia and North America. The coefficients of NTM for four regions are negative and statistically significant, meaning that the presence of NTMs has a negative impact on the imports from all selected regions. The coefficients of NTM indicate that a 1% reduction in the total number of NTMs applied by Saudi Arabia is associated with 0.43%, 0.28%, 0.35%, and 0.34% increases in import values from GCC, Asia, EU, and North America, respectively. The distance coefficients are negative and significant, as expected for all regions. Despite the existence of a trade agreement between the GCC, the effect of NTMs on Saudi imports from the GCC is larger than from other trade partners. This could be due to similarities in the economies of GCC; there is no trade in locally produced commodities, and trade transactions are mainly in re-exported goods.

6. Discussion

The results show that the per capita GDP is affecting the import value of Saudi Arabia only in the long-run and has no impact in the short-run. Generally, the response of import demand to income is positive, but, when considering the short-run and long-run effects, the findings of different studies vary, e.g., Chen [27] found that the income affects import demand in the short-run and long-run, but the effect of income is greater in the short-run than in the long-run. Khan et al. [28], however, found a long-run relationship between import demand and its income components.

Regarding the NTM impact on trade, the study results confirm the negative effect of the application of NTMs on the imports of Saudi Arabia, as it restricts import competition in the short-run and long-run. These results are consistent with the findings of many studies. For example, Cali et al. [29] proved that the effect of NTMs is worse than that of import tariffs. Yue [30] also reported a negative impact of NTMs; his findings prove that the NTMs reduce total imports and force exits. Carrère et al. [31] found that NTMs are more restrictive than the corresponding tariffs, with two thirds of the ad valorem equivalent estimates in the 25–50% range. The larger negative effect of NTMs is reported for trade between Saudi Arabia and other GCC countries because the costs associated with cross-border trade, even between well-integrated countries, were well above those that could be explained by geographic distance and traditional trade policies, as explained by Anderson and Van Wincoop [5]. Some studies, like Ghodsi et al. [15], showed that the net effects of NTMs on trade are positive, as the positive effect on the demand side compensates for the negative effect of increasing costs on the supply side. Our results dealt with only the supply side, like many other studies, e.g., Khalid and Ghafor [12], who investigated the protectionist effect of food safety standards on imports from selected developing countries. Also, Bernini et al. [17] investigated the impact of non-tariff barriers in Argentina, with a specific focus on non-automatic import licenses (NAILs) and their influence on firm exporting dynamics and employment of firms that rely on imported inputs.

7. Conclusions

Saudi Arabia applies a wide range of non-tariff measures, and their effect on trade partners has not been quantified. This study estimated a gravity model using the ARDL approach. The result of the bound test confirms the presence of cointegration between model variables and the existence of long-run and short-run effects of NTMs. The negative effect of NTMs is reported for all selected regions in the short-run and long-run, with interregional variation. The import value of Saudi Arabia is adversely affected by the application of technical and non-technical tariff measures, especially imports from the GCC region, despite the presence of trade integration. The imports from the selected regions will improve if the applied NTMs are reformed and adjusted. Therefore, it is advisable for policymakers in Saudi Arabia to adopt a more transparent policy that leads to a sustainable supply of goods and services, and, at the same time, protects the domestic markets, health, and environment. Any implementation, reform, or administration of NTMs should precisely target the market failures they are trying to correct in order to minimize the distortion costs imposed on the economy and trade. Adjustments of NTMs in textile and food sectors, transportation, and machines and electrics, with 100%, 94%, and 90% coverage ratios, respectively, and 70% of import share, may reduce restrictions and enhance trade.

Limitations of This Study

The use of aggregate panel data for the selected regions is one of the limitations of this study. Additionally, due to the unavailability of detailed data for NTMs, this study uses a dummy variable to capture their effects. Hence, we suggest that future studies quantify the effect of NTMs on trade by sectors (food and industrial products) and by individual partner countries.