 Good morning. Thank you very much. Right, so the study that I'm going to be briefly presenting this morning follows quite closely on the study that Channing just presented. But the idea now is to move on to unpacking this aid-growth relationship. So just to be clear, the majority of the studies in the literature, R site has been going on for 30, 40 years really, asks at the aggregate level whether age increases growth. This is obviously a very important question. It's important because it speaks to what we might call a first order question of should we give? Age. Equally for a poor country, should we receive age? So that aggregate question is obviously critical, Ond y dweud am berthynas iawn iddyn nhw'n mayfyn o'r gweithio gyda'i… Ac mae'n gweld eich tro, yn dweud o'r gweithio, ac mae'n dweud eich tro ar gyfer Channing, sy'n dweud o gweld eich tro o'r gweithio'r llunio'r gweithio, mae'n dweud ym mhwybwysig, mae'n dweud eich tro o'r gweithio'r gweithio'r gweithio. Felly mae'n dweud i'ch gweithio'r gweithio'r gweithio'r gweithio'r gweithio. So y mynd i'r ffordd ystod o'r ddweud yma ei fod yn ei bod yn ymwiel yn ysgolio'ch ddisigol ac yn ddweud ydych yn gallu'r ddweud o'r gweithio cyffredig, sydd ydym yn ei gael arniidd o'r ddweud. Fydde ydy'r cyffredig yn gallu'n ddweud yma mae adegau'r ddweud yn ysgolio'r ddweud oherwydd o'r llanel, ac yn ystod o'r ddweud yn ysgolio'n ddweud o'r ddweud o'r ddweud. there could be a net effect. So, it's important to identify which ones are positive, which channels might be negative for example. I think it's also important to say that growth is not the only outcome of interest. We are also typically interest in, for example, from the perspective of the Millennium Development Goals, a lot of outcomes such education, infant mortality are seen as important in themselves regardless of whether we're supporting growth or not. llawer oedd ydych chi'n cynhyrchu bod ydyn nhw, y先ant, bywyd yn ei gweithio'r gelwch arwallon yn y papur? Llyrfer felly nad oedd yw i ddweud? Mae y rhodod rhai ybydd yn gweithio'r sphel yma gyda fewn eich ymddangos ffynol oexaidd. Felly wefwnt at y peth yn cyfnod cynnyn beth ydych chi'n ymddangos cydylliannig o'r rhaid â hyrfodol i ddodgenedig. Yn ei ddagwch ar lawer o'r serfanaid yma i ddau i ffyrdd yn rhoi'r gweithon. Rydych chi'n dda i fynd i ddeithasio'r data ond yn yw'r bod yn ychydigiaeth ydy ymgyrch yn cael eu cyfrifolion o'r ddau ac yn gweld. Ond mae'r ddweud i'r ddeithasio'r ddau yn cael eu gwybodol. Ond ydych chi'n ddweud i'r ddau'r ddau ymgyrch yn cyrraedd yr adeiladau ymig o'r ddau'r gweld ymgyrch. Dyna'r ddau'r ddau'r cyfrifolion o ddau'r ddau'r ddau. Yn y cyfnod ymlaen i'r gael i'r cyffinol i'r ystyried a'r gweithio. Mae'n iawn i'r gweithio ddau, a mae'n gweithio'r gweithio'r gweithio, ac mae gennym ni'n gwahanol o'r data setiael. Mae'n gwahanol o'r gweithio o'r 1970 i 2007. Felly, yn y cwestiwch, mae'n gwahanol o'r 1970 i 2000. Ond mae'n gwahanol o'r gweithio i'r gweithio ac mae'n gwahanol o'r gweithio'r gweithio. we also serve a role in contacts with other stakeholders in the platform. Secondly, we quantify the causal impact of aid on a range of what we call intermediate outcomes. An example here is the impact of aid, for example, on education. Thirdly, we pull these two together and then identify what other transmission channels and try and basically disaggregate the aid growth relationship that we find. ...fyrraedd o'n ffordd o gyhoedl o'r problemau wneud oherwydd weithio y ddiddordebol sy'n gwneud y ffordd... ...y byddol ynmarhwg pethau gyda'r oedau ddiddordebol oherwydd os y gallan fod argymeth sy'n byddai'n ffordd... ...'i codi gwneudio ffordd o'r ddiddordebol y ddiddordebol. Efallai eraill gyrraedd, efallai'r gyrraedd Sulwyr yn y sydd wedi'i debyg o hynny... Felly, ond mae'r gwaith ymlaen i fyddiadau, ond mae'n llawer o'r brifau am gyfer yswod rhan, mae'r bwysig yn ymddangos i ddim yn dweud. Rydyn ni'n credu'n gweithio cyfnod, mae ddim y ddim yn dweud, mae'n ddim yn ddim yn gweithio'r ddim yn dweud, mennydd yn y bwysig o'r dyfodol, a diwylo'r ysbryd agrodigol, Felly, we need to be able to disentangle this reverse causality that might be going on. And that's why we need to use some techniques, essentially, to deal with that. And we also overall take inspiration from the latest paper that Channing presented. And it's a long cross-section approach. So we take lots of different countries and look at their behaviour over a period of time. So what are the results? Now, this is what we're really here for. Oh, this is difficult. Okay. Strain my neck. What this table very simply presents is simply a number of different final and intermediate outcomes. And obviously the outcomes are here we look at individually. And then the next column is a baseline. And that's just the median of all the countries over the period of time 1970 to 2007. So this 1.7 here is the median GDP per capita growth rate of the countries in the sample. So similarly, the median poverty rate is 21.7. And then based on our results, we ask if all these countries were given an additional 25 US dollars of aid per capita each year. So if everyone in all these countries were given this amount of aid, what would the outcome be? So this is really the difference between the baseline and the effect of this additional aid. So that's given in this final column here. And in each case, we find a positive, well, a contribution that goes in line with what we would expect. So with respect to GDP per capita, the additional growth is about 0.3 percentage points. With respect to poverty, we find poverty goes down by a small number of percentage points. Interestingly, agriculture as a percentage GDP goes down. And one might question whether this is a good result or a positive result. But typically we tend to find that in the process of structural transformation of development, that agriculture does tend to decline as a percentage GDP. All this means is that remember that we find that there is economic growth. So it just means that other sectors are growing faster than agriculture. We also find that investment is higher. And we find that average years of schooling and life expectancy at birth are also higher. So what we find, therefore, is that aid not only supports growth, it also has a positive contribution across a range of other outcomes. Now this is important because it kind of builds on the consistency of this. This aggregate result, it makes sense that we find positive contributions elsewhere in the economic system. At the same time, it's important to notice that these are relatively modest contributions. So this builds on and reinforces the point made by Channing that growth is a long-term cumulative process and changes take a long time to come through. And aid is only one part of supporting that process. Now we look at the disaggregation of the transmission channels. And I think the main results are actually right at the top here. And to make it simple, we focus on three main transmission channels. The transmission channel from aid to investment. By investment, we mean investment in physical capital like infrastructure, roads, et cetera, et cetera, to growth. We look at the impact or the transmission channel going from aid to education to growth. And the transmission channel from aid to health to growth. Now what we find is that the first and the last two essentially explain all of the contribution of aid to growth. So the aid investment channel accounts for around 75% of the aggregate impact of aid on growth. The aid to health channel explains around 25% of the aggregate impact of aid on growth. Now the big question is what's happening to education? Well what we do find here, which is interesting, is that there is a positive impact of aid on education. We saw that in the previous slide. In other words, that countries that receive more aid tend to show better educational outcomes. Good. The problem is the link from education to growth. And that's what we don't find a positive effect. In fact, the table below gives you the numbers and it's potentially a little bit difficult to understand. But this is the contribution of education to growth, which is essentially not distinguishable from zero. Now this can create a host of questions. But let me tell you that this is not the first time that this has been found in the literature, in the research. This is actually an area which is highly debated. There have been a number of papers by economists much more famous than I am that say where has all the education gone? So there is a lot of debate around the contribution of education to growth. One of the reasons that we might be seeing this, but not able to find a positive effect, is simply because the data we use on education is very noisy and it doesn't reflect differences in educational quality. So a person can go to school for four years, but that four years or ten years means one thing in one country and it might mean something very different in another country. But all we say is how many years did you go to school? Ten years. And we're kind of assuming that this is the same quality, but in fact it's not. So when you find or when there are problems in the data, estimates quickly go to zero. So that could be an explanation for this effect, but we're not sure. So this is the table there. I won't go into the details, but that's just the disaggregation of the different effects. So to sum up. So supporting the first presentation, we find there is a highly consistent and coherent pattern of results across what we heard about the microevidence and now we've looked at the mesoevidence, that's the intermediate level and the macroevidence. It does add up. Once again, it's the cumulative impact of aid that's important. There are no quick wins. What we find from these results as well is that the internal rate of return from aid to growth is around 16%. So that is a number which is substantial, reasonable, and this kind of a rate of return is often used in the investment literature as to whether an investment is a good deal and 10% is often taken as a kind of a benchmark. So 16% is positive. Probably one of the main results is the disaggregation and we find that physical investment and human capital are the main principal channels through which aid supports growth. But just to remind us all, aid also is found to support other outcomes. It's just that there is a more ambiguous relationship between these other outcomes and growth, such as education. Thank you very much. Could I just ask you whether you said it's a data problem and very much discussed this link between education contributes to growth? Do you think during your research you will find sort of better data or more careful data so you can establish some connection because it's a bit surprising that education does not contribute to growth? Yes, I think my immediate reaction would be that there is better data out there. The problem is that it doesn't really cover developing countries. So that's one of the difficults we find, for example, I mean we're interested in the impact of aid on largely developing or low income countries and it's those countries where we tend to have the worst data in terms of educational quality. But it is something that we perhaps should dig into a bit further.