Modeling is not an easy task to simulate the reality. In sport performance, modeling the human body in a particular event still lack many variable and has many assumptions. This deviate the model from the reality. In all kind of modeling you cant collect all variable as an input in one model.

In my opinion, Aid could have positive impact on growth in developing countries depending on the use of the recipient country. for the method applied the majority utilize the panel with instrumental variable: in the model growth and aid are endogenous.

Another remark: the use of ODA may be not appropriate in several papers because of the surevaluation of assistance funds. the bulk of recent papers used EDA: Effective development assistance: Kindly take a look on the work of Serven et al (1998) about calculating EDA.

http://econ.worldbank.org/WBSITE/EXTERNAL/EXTDEC/EXTRESEARCH/0,,contentMDK:20827466~pagePK:64214825~piPK:64214943~theSitePK:469382,00.html

I believe that the reason for this is that so many other relevant variables relating to human behavior were not and could not be possibly included in the models, or could not be consistently measured.

And the over-expectation for results compared to the investment made.

ABSORPTION CAPACITY: Less developed countries generally have poor institutional infrastructure and the governments also have problems with fiscal and monetary management. Many of these countries generally have lower GDP and are partially dependent on foreign aids. Despite continued financial support from outside, the countries remain poor, i.e. low GDP---Somalia, or Haiti, for instance. On the other hand, in a more developed economies, the foreign aids may come in different forms, other than cash or cash equivalence assistance. Foreign aid to help build universities to train personnel that would later become productive in the economy, for instance, would be difficult to measure if it takes 4 - 6 years to train productive professionals.

TYPES OF FOREIGN AIDS: the type of aid may also affect the outcome. Some aids may not be monetary, but has monetary value. The sending of food assistance to drought stricken country or sending medical aids to an epidemic area has monetary value and is counted as foreign aid. However, this type of aid would most likely not result in the increase in the GDP of that country.

DATA: for all countries receiving foreign aids and their GDP report up to 2014, see the IMF’s annual report:

http://www.imf.org/external/pubs/ft/weo/2014/01/

META ANALYSIS: meta analysis in this field is difficult because researchers may have different research objective and, thus, each looks at the effect of foreign aids on GDP in different perspectives. While all these research have sound models and employed sound methodologies, when comparing their results we must look each articles time frame of the data and types of economies they examined.

Official development assistance to some countries did not work probably due to , at least , 2 reasons: 1) The receiving country has not been investigated well by the donors (who sometimes search & judge from their "ivory towers" or depend on reports that are inaccurate). 2) There are countries which cannot improve whatever aid is given simply because they contain professional thieves who can overcome any mechanism of supervision or watching!

There are at least three problems at this level: 1 Choice of method (ie not taking into account the problem of reverse causality, among others). 2. measurement errors. 

 3. Foreign aids unlike remittances go through complicated administrative channels , undermining their effectiveness (corruption ...).

It is therefore difficult to isolate the impact of aid on growth and development.

Hello,

Economic growth is not changing. But other factors are changing or variable? Perhaps such aid are impacted in ways that are not being measured or studied. Qualitative studies can complement such findings on the impact of such aid. They are more likely to show things that go unnoticed by the quantitative analysis. There unwanted but also real that should be taken into consideration as the deviation and corruption variables, for example, very well remembered by our colleague Nizar Kill, but that can have a big impact. The volume of these deviations in developing countries are often not negligible. Sometimes, too, the explanatory models we create in our heads do not express what actually occurs. Then it's also good to examine the assumptions of the model, the variables in question, etc. It is worth reviewing to explanatory theory about the impact of such aid. As I said above perhaps the real impacts are being imagined impacts of different (theoretical explanatory model).

Regards,

S.

No foreign aid is only economical; by descriptions of experiences I read, the equality and accord between so-called helpers and helped people in  the common endeavor has the principal impact. I quote the canadian implications in Afghanistan : support to microcredit programs seems to have better economic impact than mere money to invest in megaprojects.

Positive consequences depend on the degree of integration of the aid policy in the hole economic activity of the "helped" country. It is also affected by the balance of power between both countries. These indirect effects may affect the econometric results.

http://www.ecrg.ro/files/p2014.4%281%2912y4.pdf

Please can you provide references on the 'paradoxical results' you indicated?

In my limited experience, I would question the quality of information used to inform models. As the ultimate aim is to benchmark activities (to make best use of limited foreign investment resources), a suggestion to international funders would be to agree a standardised assessment template (agreed by a consensus group) to evaluate the impact of foreign investment and associated costs. Making this data available to the research community would improving visibility on activities, information quality and allow examination of cost efficiencies in achieving the intended goals of investments.

There is no contradiction:

1. ODA is applied to low return investment or a longer term is needed for fruits to come along and by that time the ODA is not there.

2. ODA has strings attached and many times is given in support of governance, democracy, institutions,..., and people do not eat those things. Production is the ones that generated employment and economic growth.

3. ODA is there today and is not there tomorrow. In other words, it depends on external factors and continuity is not guaranteed.

4. ODA is so little relative to other consumption and investment spending that its impact is next to nil, in many cases. Moreover, ODA is applied to specific projects and lacks macro-impact.

5. If the econometric analysis involves many countries, then the results are near meaningless: (5.1) The WB, the IMF and others put the countries in bunches and then they applied a not thinking software to get the results. The final output is semi-useless because countries in Africa are not compared to countries in Latin America and the Caribbean or versus the industrial countries; (5.2) As Solow uses to say: the economic growth model was meant or created for industrial countries, and there was no intention to put together a bunch of countries in the equation; (5.3) but more important for a country or even for similar countries you have to distangle (disaggregate) economic growth as related to sectoral growth (Agriculture, Manufacture, Services-Tourism). As you can see is not econometrics, judgement and analysis are missing pieces.

Choice of  variables. E.g. aid goes to countries with pleasant climate and conditions for donor staff. It goes to countries like Zambia which look big on the map, rather than to smaller ones with the same population, like Malawi, it is given for political reasons. Tanzania gets a lot of aid because of the game reserves. Rationally one might give more aid to the 'basket cases' which are least able to use it effectively. And so on. All of which means that an analysis assuming these away, assuming all countries are identical except for the variables in the model (those for which data happens to be available) will come up with 'perverse' results. And it may be that aid is harmful: compare Rhodesia under UDI with most African countries in the last 50 years. As Vanegas says, the econometric results are meaningless,

How do GDP and particularly the Gini coefficient effect the likely "aid" given to countries? A very worthy investigation although I'm not at all conversant with the maths and statistics